Lorem (Link Bibliography)

“Lorem” links:

  1. Design

  2. 2019-02-18-lecun-isscc-talk-deeplearninghardwarepastpresentandfuture.pdf#page=60: “Deep Learning Hardware: Past, Present, & Future”⁠, Yann LeCun

  3. https://commons.wikimedia.org/wiki/File:Japanese_crest_Tsuki_ni_Hoshi.svg

  4. Sidenotes

  5. ⁠, Akiva Miller, Ilene Miller (1998):

    [extensive satire of Jewish Halakha, describing the rulings of the Kringler Rav on the proper methods of preparing for Xmas, the tree, decorations, gifts, conduct at office parties and the festive meal, rulings on the existence of Santa Claus (“yes”), miscellaneous customs, the Christmas Havdalah ceremony, a transcript of the Hagada to recite on Xmas, and miscellaneous Xmas songs like ⁠.]



    1 This contrasts sharply with Shabbos, for the mitzva of honoring Shabbos applies all week long. For example, if one finds a particularly good food during the week, one should save it for Shabbos even though it is now only Sunday and Shabbos is a week away. However, Xmas preparations may not begin too far in advance, in order to fulfill the dictum, “It’s beginning to look a lot like Xmas.”
    2This is because of the principle that two festive occasions should not be mixed into each other. Note the decree of the great R. H. Macy, who established that Santa Claus may not appear in the Thanksgiving Day parade until after all the other floats have passed.
    3 There are some who begin preparing for Xmas as early as Halloween. This is wrong, and they will be called upon to account for their evil ways.
    4 Such as setting up the Xmas tree (some say even buying one), or playing holiday music on the Muzak.
    5 Such as buying gifts or buying the Xmas dinner turkey. Cooking the turkey may not be done before Thanksgiving because it will appear to be a Thanksgiving turkey.

  6. ⁠, Cendri Hutcherson, Constantine Sharpinskyi, Michael Varnum, Amanda Rotella, Alexandra Wormley, Louis Tay, Igor Grossmann (2021-03-26):

    Effective management of global crises relies on expert judgment of their societal effects. How accurate are such judgments?

    In the spring of 2020, we asked behavioral scientists (n = 717) and lay Americans (n = 394) to make predictions about COVID-19 pandemic-related societal change across social and psychological domains. Six months later we obtained retrospective assessments for the same domains (Nscientists = 270; NlayPeople = 411). Scientists and lay people were equally inaccurate in judging COVID’s impact, both in prospective predictions and retrospective assessments. Across studies and samples, estimates of the magnitude of change were off by more than 20% and less than half of participants accurately predicted the direction of changes. Critically, these insights go against public perceptions of behavioral scientists’ ability to forecast such changes (n = 203): behavioral scientists were considered most likely to accurately predict societal change and most sought after for recommendations across a wide range of professions.

    Taken together, we find that behavioral scientists and lay people fared poorly at predicting the societal consequences of the pandemic and misperceive what effects it may have already had.

    Figure 1: Prospective (April 2020) and retrospective (October/​​​​November 2020) judgment of societal change along with objective markers (dotted line) behavioral scientists and lay people. We included behavioral scientists from both Studies 1 and 2 because their forecasts were largely the same (see Table S10). Error bars indicate 95% confidence interval.
  7. /

  8. #matt-lakeman-2020-peepshow

  9. Subscripts

  10. 2019-radford-figure4-gpt2validationloss.png

  11. 2020-brown-figure31-gpt3scaling.png

  12. https://arxiv.org/pdf/2005.14165.pdf#org=openai&page=46

  13. ⁠, Dario Amodei, Danny Hernandez, Girish Sastry, Jack Clark, Greg Brockman, Ilya Sutskever (2018-05-26):

    [Further reading: “Parameter Counts In Machine Learning” (2021-06-19), Akronomicon leaderboard⁠.] We’re releasing an analysis showing that since 2012, the amount of compute used in the largest AI training runs has been increasing exponentially with a 3.4-month doubling time (by comparison, Moore’s Law had a 2-year doubling period). Since 2012, this metric has grown by more than 300,000× (a 2-year doubling period would yield only a 7× increase). Improvements in compute have been a key component of AI progress, so as long as this trend continues, it’s worth preparing for the implications of systems far outside today’s capabilities.

    Three factors drive the advance of AI: algorithmic innovation, data (which can be either supervised data or interactive environments), and the amount of compute available for training. Algorithmic innovation and data are difficult to track, but compute is unusually quantifiable, providing an opportunity to measure one input to AI progress. Of course, the use of massive compute sometimes just exposes the shortcomings of our current algorithms. But at least within many current domains, more compute seems to lead predictably to ⁠, and is often complementary to algorithmic advances…The trend represents an increase by roughly a factor of 10 each year. It’s been partly driven by custom hardware that allows more operations to be performed per second for a given price (GPUs and ), but it’s been primarily propelled by researchers repeatedly finding ways to use more chips in parallel and being willing to pay the economic cost of doing so.

    AlexNet to AlphaGo Zero: A 300,000× Increase in Compute. (The total amount of compute, in petaflop/​​​​s-days, used to train selected results that are relatively well known, used a lot of compute for their time, and gave enough information to estimate the compute used.)

    Eras: Looking at the graph we can roughly see 4 distinct eras:

    1. Before 2012: It was uncommon to use GPUs for ML, making any of the results in the graph difficult to achieve.
    2. 2012–2014: Infrastructure to train on many GPUs was uncommon, so most results used 1–8 GPUs rated at 1–2 TFLOPS for a total of 0.001–0.1 pfs-days.
    3. 2014–2016: Large-scale results used 10–100 GPUs rated at 5–10 TFLOPS, resulting in 0.1–10 pfs-days. Diminishing returns on data parallelism meant that larger training runs had limited value.
    4. 2016–2017: Approaches that allow greater algorithmic parallelism such as huge batch sizes, architecture search, and expert iteration, along with specialized hardware such as TPUs and faster interconnects, have greatly increased these limits, at least for some applications.

    ⁠/​​​​ is the most visible public example of massive algorithmic parallelism, but many other applications at this scale are now algorithmically possible, and may already be happening in a production context.

    Addendum: Compute used in older headline results (2019-11-07)

    We’ve updated our analysis with data that span 1959 to 2012. Looking at the data as a whole, we clearly see two distinct eras of training AI systems in terms of compute-usage: (a) a first era, from 1959 to 2012, which is defined by results that roughly track Moore’s law, and (b) the modern era, from 2012 to now, of results using computational power that substantially outpaces macro trends. The history of investment in AI broadly is usually told as a story of booms and busts, but we don’t see that reflected in the historical trend of compute used by learning systems. It seems that AI winters and periods of excitement had a small effect on compute used to train models over the last half-century.

    Two Distinct Eras of Compute Usage in Training AI Systems

    Starting from the perceptron in 1959, we see a ~2-year doubling time for the compute used in these historical results—with a 3.4-month doubling time starting in ~2012. It’s difficult to draw a strong conclusion from this data alone, but we believe that this trend is probably due to a combination of the limits on the amount of compute that was possible to use for those results and the willingness to spend on scaling up experiments. [For one vivid account of the history of computing in AI in this period, see the “False Start” section in ⁠.]

  14. image-focus.js

  15. 1962-teller-thelegacyofhiroshima.pdf: “The Legacy of Hiroshima”⁠, Edward Teller, Allen Brown

  16. ⁠, Morgan McGuire ():

    Admonitions are small break-out boxes with notes, tips, warnings, etc. for the reader. They begin with a title line of a pattern of three exclamation marks, an optional CSS class, and an optional title. All following lines that are indented at least three spaces are included in the body, which may include multiple paragraphs. The default stylesheet provides classes for “note” (default), “tip”, “warning”, and “error”.

  17. ⁠, Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Mark Chen, Rewon Child, Vedant Misra, Pamela Mishkin, Gretchen Krueger, Sandhini Agarwal, Ilya Sutskever (2021-01-05):

    [Paper: ⁠, Ramesh et al 2021. Re-implementation: DALL·E Mini (writeup). cf ⁠, ⁠. Availability through OA API still planned as of 2021-09-05.] DALL·E is a 12-billion parameter version of trained to generate images from text descriptions, using a dataset of text-image pairs. We’ve found that it has a diverse set of capabilities, including creating anthropomorphized versions of animals and objects, combining unrelated concepts in plausible ways, rendering text, and applying transformations to existing images.

    GPT-3 showed that language can be used to instruct a large neural network to perform a variety of text generation tasks. showed that the same type of neural network can also be used to generate images with high fidelity. [iGPT is another answer to the question of “how do we do images autoregressively, but not at the exorbitant cost of generating pixels 1 by 1?”; iGPT uses ‘super pixels’ & very small images, while DALL·E uses VAE ‘tokens’ corresponding roughly to small squares so the token sequence is relatively small, where the VAE does the actual compilation to raw pixels.] we extend these findings to show that manipulating visual concepts through language is now within reach.

    [3 DALL·E prompts: “an armchair in the shape of an avocado…” · “a store front that has the word ‘openai’ written on it…” · “the exact same cat on the top as a sketch on the bottom”]

    DALL·E’s vocabulary has tokens for both text and image concepts. Specifically, each image caption is represented using a maximum of 256 BPE-encoded tokens with a vocabulary size of 16384, and the image is represented using 1024 tokens with a vocabulary size of 8192. The images are preprocessed to 256×256 resolution during training. Similar to VQ-VAE, each image is compressed to a 32×32 grid of discrete codes using a discrete VAE1011 that we pretrained using a continuous relaxation. We found that training using the relaxation obviates the need for an explicit codebook, loss, or tricks like dead code revival, and can scale up to large vocabulary sizes.

    Capabilities: We find that DALL·E is able to create plausible images for a great variety of sentences that explore the compositional structure of language. We illustrate this using a series of interactive visuals in the next section. The samples shown for each caption in the visuals are obtained by taking the top 32 of 512 after reranking with ⁠, but we do not use any manual cherry-picking, aside from the thumbnails and standalone images that appear outside.

    1. Controlling attributes: We test DALL·E’s ability to modify several of an object’s attributes, as well as the number of times that it appears.

    2. Drawing multiple objects

    3. Visualizing perspective and three-dimensionality

    4. Visualizing internal and external structure

    5. Inferring contextual details

      We find that DALL·E is able to render the same scene in a variety of different styles, and can adapt the lighting, shadows, and environment based on the time of day or season: “a … of a capybara sitting in a field at sunrise”

    …With varying degrees of reliability, DALL·E provides access to a subset of the capabilities of a 3D rendering engine via natural language. It can independently control the attributes of a small number of objects, and to a limited extent, how many there are, and how they are arranged with respect to one another. It can also control the location and angle from which a scene is rendered, and can generate known objects in compliance with precise specifications of angle and lighting conditions.

    Zero-shot visual reasoning: GPT-3 can be instructed to perform many kinds of tasks solely from a description and a cue to generate the answer supplied in its prompt, without any additional training. For example, when prompted with the phrase “here is the sentence ‘a person walking his dog in the park’ translated into French:”, GPT-3 answers un homme qui promène son chien dans le parc.” This capability is called zero-shot reasoning. We find that DALL·E extends this capability to the visual domain, and is able to perform several kinds of image-to-image translation tasks when prompted in the right way. [See also ⁠.]

    We did not anticipate that this capability would emerge, and made no modifications to the neural network or training procedure to encourage it. Motivated by these results, we measure DALL·E’s aptitude for analogical reasoning problems by testing it on Raven’s Progressive Matrices, a visual IQ test that saw widespread use in the 20th century. Rather than treating the IQ test a multiple-choice problem as originally intended, we ask DALL·E to complete the bottom-right corner of each image using argmax sampling, and consider its completion to be correct if it is a close visual match to the original. DALL·E is often able to solve matrices that involve continuing simple patterns or basic geometric reasoning, such as those in sets B and C. It is sometimes able to solve matrices that involve recognizing permutations and applying boolean operations, such as those in set D. The instances in set E tend to be the most difficult, and DALL·E gets almost none of them correct. For each of the sets, we measure DALL·E’s performance on both the original images, and the images with the colors inverted. The inversion of colors should pose no additional difficulty for a human, yet does generally impair DALL·E’s performance, suggesting its capabilities may be brittle in unexpected ways.

  18. sidenotes.js

  19. tablesorter.js

  20. ⁠, Matt Lakeman (2020-03-21):

    [“A warning against assuming the immense emotional and moral responsibilities that come with caring for a dog. Can an owned animal have a good life?” Imagine that you, a human, were kidnapped by aliens at birth and given an approximation of a dog’s life, and a good dog’s life at that. Ignore the subservience, dependence on a superior life form, and all the other psychological aspects of being owned and just focus on how you would feel about your material conditions. Would you want this life?" (7,700 words)" —The Browser summary

    Meditation on pet ownership. What is the morality of keeping a mentally and physically crippled animal, particularly in an urban apartment where it cannot exercise its natural urges or get adequate exercise/​​​​stimulation? The ‘cute’ behavior of a dog, so appealing to so many, is, regarded more cynically, indicative of severe pathology and dependency, a Stockholm syndrome; aside from the effects on the slave, what are the effects on the master? At least a cat’s trust and affection has to be earned; what should we think of humans who love the pathetically unconditional love of a dog?]

  21. https://www.gwern.net/reviews/Cat-Sense

  22. ⁠, Matt Lakeman (2019-12-08):

    [Review/​​​​summary of Andrew Roberts’s 2014 Napoleon: A Life⁠.]

    had one of the most accomplished, divisive, big lives of any person in history, which reshaping the way we think about war, politics, revolution, culture, law, religion, and so much more in a mere 52 years. Any one of those elements could (and has) been isolated and made into a massive tome on its own.

    So I just set out to describe and analyze all of the things I found most interesting about the man. This includes a summary of his entire life, his personality quirks, unusual events, driving beliefs, notable skills, and more. If there is an over-arching theme to be found, it’s my amazement at how an extraordinarily competent and risk-tolerant individual lived his life up to the greatest heights only to come tumbling back down to earth.

  23. ⁠, Matt Lakeman (2019-09-03):

    [Review of Lenora Chu’s 2017 memoir Little Soldiers: An American Boy, a Chinese School, and the Global Race to Achieve⁠, a memoir by a Chinese-American journalist married to a Jewish NPR journalist of subjecting their child to elite Chinese education in Shanghai. Chu naively credits the brutal rigor and training of the Chinese education system for the stellar results on international comparisons like PISA (the truth is likely otherwise).]

    That’s why this was pure nightmare fuel for me. It’s a non-fiction account of an ethnically-Chinese, American-born woman following her multi-racial child through the Chinese school system in Shanghai. While we complain about our soft, liberal, decadent school experiences in America or Europe, tens of millions of Chinese kids are subjected to a school structure that seems purposefully designed to make everyone as miserable as humanly possible. Or at least that was my take-away. Lenora Chu has a kinder perspective on the system. Mostly.

    …Why do students and parents put up with all this? Are they just brainwashed by an authoritarian culture and government? Probably yes, at least to some degree. But there’s also a rational calculus at play…The modern manifestation of the Imperial Exam is the ⁠. This single test given during the last year of secondary school is the sole determinant of the vast majority of Chinese students’ abilities to get into a domestic university. Its subject matter is certainly more practical than the old exams—math, Chinese literature, and English—though probably still fairly esoteric by Bryan Caplan’s standards.

    Lenora stresses that though the gaokao is the flagship mega-test that everyone focuses on, basically all of Chinese education is a series of gaokaos. Students and parents believe that every homework assignment, quiz, and class test serves as a signal of a student’s ability in a giant country-wide competition to reach the top echelon of Chinese society. Every failure, no matter how small, sets a student back and jeopardizes his entire life. This is why one of Lenora’s friends enrolled his three-year-old son in a pre-MBA program. It’s also why virtually all of Rainey’s classmates’ parents thought Lenora was a horrible mother for giving Rainey “idle time” on the weekends instead of enrolling him in Chinese writing classes, English classes, piano classes, algebra classes, etc. All the parents believed that by getting their kids into these programs at this stage (barely post-toddler), they were getting a head-start, and that head-start would carry into primary school, then secondary school, then university, then work, then life. It was all about getting ahead.

    Lenora did her best to avoid the rat race, but found that it was too all-encompassing. At Soong Qing Ling, Lenora dreaded seeing the “Big Board”—a bulletin board hung up outside Rainey’s classroom where his teacher posted constant updates of students’ stats. Some of these stats were relatively innocuous—like the height and weight of every student—while other stats ranged from shoe size and blood type, to individual teacher’s comments on instrument playing (apparently Rainey had no sense of rhythm). Naturally, Lenora didn’t care about any of this stuff, but the other parents did. She watched mothers gloat over the physical prowess of their three-year-old children, take shots at each other over public admonishments, and generally make mountains out of utterly benign molehills. Lenora’s enduring shame was that Rainey was bad at playing the recorder. Despite getting stern warnings from the teacher and other parents, Rainey never improved, and eventually he was the only student left out of the class’s recorder concert in front of all the parents.

    Lenora realized that, again, there was a method to the madness. Soong Qing Ling was training students and parents to compete. A three-year-old who learns to compete over public displays of height, recorders, and blood types, will learn to compete over grades in the future. Parents who are ready to fight the same battles will drive their children into the fray for years to come. Everyone learned that they were at war, or rather, in a battle royale. Only the very best survive, and pre-school at Soong Qing Ling was just the beginning.

    …I guess I couldn’t help but come away from the book with a sense of depressing apathy. There’s so much effort put into education by well-meaning people, but a lot of it just seems to make children and parents hate everything. China apparently spent 2,700 years perfecting its education system, and this is the result… students being force-fed, parents shamed into kowtowing, and teachers corruptly abusing their power. Yes, there are good people in the system to, and apparently the children learn something and it does give some sort of order to society, but I really, really wish they didn’t have to this way.

  24. ⁠, Matt Lakeman (2019-10-27):

    ⁠, a British TV series running from 2003 to 2015, starring David Mitchell and Robert Webb as a pair of miserable, co-dependent roommates living in Croydon, London, is the most realistic portrayal of evil I have ever seen.

    …We see this all not just by watching Mark and Jez go about their day-to-day lives, but by hearing their inner thoughts through voice-over monologues, which more often than not, reveal their actions and words as either cynical attempts to avoid facing their own failings, or desperate lies to obscure their true intentions, goals, and personalities.

    This is what makes Peep Show so brilliant. It doesn’t just portray evil realistically, it portrays the root of evil realistically. Mark and Jeremy cause bad things to happen to their acquaintances, co-workers, friends, loved ones, family members, and most of all, themselves, because they are consumed by their vices. Not just the classic vices like gluttony and lust, but cowardice, evasion, hypocrisy, and apathy, all born from a rarely acknowledged, yet omnipresent self-loathing. These are vices that aren’t loudly announced by violent or easily identified in scary individuals, but vices that sneak up on ordinary people, latch on to their psyches, and take over their lives.

    Also, it’s one of the funniest TV shows I’ve ever seen.

  25. ⁠, Matt Lakeman (2018-12-31):

    I’ve written a couple of book summaries on here over the past few months, and this one for will be the most difficult. ’s autobiography is a sociological summary of Appalachian American culture, and by extension the culture of poverty across America, which uses his own life as a case study. The book is basically a series of linked anecdotes with only occasional introspections thrown in, so I’ll try my best to lay out Vance’s story, and integrate his claims and arguments.

    …You know that classic Republican straw man about poor people? It goes something like—

    “In the glorious modern American capitalist economy, all people can pick themselves up by their bootstraps and make a good living if they really want to. The only way to fail is to not try hard enough. Poor people are all lazy loafers who would rather take drugs, rack up illegitimate children, and become welfare queens, than work an honest day in their lives. It’s their own damn fault they’re poor.”

    Vance argues that this straw man is basically true.

    Yes, of course it’s more complicated than that. There are external factors at play that makes the lives of his fellow hillbillies in Appalachia worse, like the collapse of American industrialism. But underlying the depressed economies, high unemployment, underfunded schools, and shoddy welfare networks, are simply a lot of bad decisions made on an individual level…Only a very select few hillbillies “make it” in the sense of achieving a stable, middle-class lifestyle. J. D. Vance is one of those few. He starts off the book saying that he feels ridiculous writing a memoir because his “greatest accomplishment” to date was graduating from Yale Law School. Yet, as he walks the reader through his life, it becomes more and more apparent just how amazing that feat is…I was aware of all these stereotypes before reading the book, but seeing them so fully fleshed out really brought home how scary it is. These people probably aren’t evil… but a lot of them are kind of bad. Or at least foolish. Or at least make really stupid decisions all the time. Somehow, that’s even scarier than being evil, or at least it’s harder to fix.

    …Vance consistently stresses that by raw material standards, nobody in Middletown was doing that badly. Yet they were miserable, depressed, addicted, and hopeless anyway.

    For instance, when Mom was with her first husband, the toothless hillbilly guy, they could be considered solidly middle-class. Mom was a nurse, her husband was a truck driver, and together they made over $100K per year with two kids in a low-cost-of-living region of America. And yet financial problems were always one of the biggest triggers of family screaming matches. They were deeply in debt because both Mom and the husband bought multiple new cars per year, they ate out every day instead of cooking, and they purchased a below-ground swimming pool. The house was already mortgaged, but was falling into disrepair due to lack of upkeep, while they repeatedly crashed new cars, and burned through meager savings with credit card fees. Vance’s family could have been fine. His parents could have lived comfortably, had good savings, and started a college fund. And maybe if they did, the stress wouldn’t have driven Mom and husband to break up, and Mom wouldn’t have turned to drugs, etc. But it didn’t turn out that way.

    Throughout the book, I had a question that I wished Vance would have answered directly. Are hillbilly values the problem, or hypocrisy against these values?

  26. ⁠, Matt Lakeman (2019-06-10):

    [Personal memoir of growing up in the rural US northeast and losing a friend to heroin overdosing. Despite living in a stable and relatively well-off white middle-class family, the friend ‘Jack’ had always suffered health problems and severe social anxiety, especially compared to his accomplished popular younger brother. Jack was never truly happy, and clashed with his brother, who resented his problems and the drain on parents. In high school, Jack gravitated to a group of bad peers who began drug use, existing in a constant malaise. A chance injury and painkiller prescription led to an addiction, and then heroin. His parents invested enormous amounts of effort into rehab and monitoring Jack and trying to get him launched on some sort of real higher education and career, if only a trade, but Jack was uninterested and kept returning to drugs in between endless video game playing. This destroyed the family finances & relationships.]

    I’m a libertarian who thinks all drugs should be legalized, including heroin. But I have to admit that learning what Jack’s addiction did to his family made me understand the “Drug Warrior” perspective better. Unless an addict has no social connections whatsoever, his addiction will hurt others. The stronger the connections, the worse the pain. If the supporting friends and family members hold on tightly enough, it will destroy them. Derrick described the five years of being with Jack through his addiction until his death as a “living hell.”

    To start with, fighting addiction costs money. Jack’s family was solidly middle-class, with his father pulling in enough money alone for the mother not to work while affording a nice home, comfortable day-to-day life, and the occasional vacation. They were decently well-off, but not enough to sustain the hit of $150K+ of rehab costs. I noticed some of the effects from afar but didn’t get the full picture until after Jack died. First the family stopped going on vacations, then the mom got part-time work (which wasn’t easy while trying to keep Jack in Lockdown), then the father worked longer hours, and eventually they were draining their retirement funds and mortgaging their house. But monetary costs were nothing compared to the emotional toll. How happy can you really be on a day-to-day basis when you come home to where your heroin-addicted son or brother lives? Jack’s parents basically lost their lives. Every single day, every single minute, was oriented around Jack. They always had to know where he was, what he was doing, when his next Narcotics Anonymous meeting was, if they could afford that therapist, etc. The father no longer worked to build college and retirement funds, but to pay off debts. The mother didn’t stay home to take care of the house and kids, but to keep her son alive. Then there was the lying…The fighting became worse than ever. They weren’t physical anymore, not while Jack deteriorated and Derrick bulked up. Yet they were more vicious than ever. More personal…For years before then, Derrick’s life had inexorably been consumed by Jack. The instant Derrick showed his parents Jack’s track marks, his childhood ended. Jack became a black hole at the center of the family which sucked everything in. Money, energy, time, and attention only flowed one way. Derrick stopped being another son and was repurposed as an asset to be employed by his parents for Jack’s sake.

    …For me, the scariest part of learning Jack’s full story was realizing that he may have been acting rationally. I’m not saying that being a heroin addict is rational, and I’m not saying that Jack made good choices, especially not given the emotional carnage left in his wake, but… I think I understand why he kept going back to the drugs…I think everyone is aware of these shitty parts of life. But almost everyone is also aware of the good parts. Family, friends, and loved ones reflect our values and fuel our lives. Hobbies, passions, and maybe even work are outlets for our virtues that convert effort and inspiration into rewards. It’s not easy, but we all fight to make the good parts as big as possible while minimizing, mitigating, or maybe even ignoring the bad parts. I don’t think Jack was ever aware of the good parts. And I think his bad parts were intrinsically worse than most people’s…Jack was painfully aware that his future options were, “be a complete loser”, or “be a complete loser who feels really really good for a few hours every day.” He chose the latter.

    …One day, when Jack was 23-years-old, his parents left the house together to see a movie. It was the first time they had gone out together without Jack in six months.. The parents came home with a cheeseburger for Jack, and they found him in his room, passed out in his own vomit on his bed. His mother called 911 while his father tried to resuscitate him, but Jack was already dead. His cause of death was an overdose, though it’s unclear whether he accidentally took too much or hit a bad batch. After the wake and funeral and shock, Derrick admitted that he felt relief. It was finally over.

  27. ⁠, Matt Lakeman (2020-04-27):

    [“Admirably lucid revisiting of Enron’s metamorphosis from a pipeline company into a derivatives trading-house that booked billions in paper profits before collapsing.” The Enron story displays the potentially distortionary impact of high intelligence on moral decision-making. It lends evidence to the notion that extremely intelligent people can be subtly incentivised to be (systematically) dishonest because their intelligence lowers the cost and raises the potential benefits of circumventing rules." —The Browser summary

    What, in a nutshell, was the Enron fraud? Like a tech startup, Enron had a vision of creating many new markets by upfront investments; to achieve this, which was in fact often a viable business strategy and had worked before, it needed debt-financing and to look like a logistics company with stable lucrative locked-in long-term profits, though its profits increasingly actually came from volatile unreliable financial trading. From this pressure and the need to keep up appearances to avoid switching horses in mid-stream before projects could pay off, a house of cards built up, deviance was normalized, and it slowly slid into an enormous financial fraud with few people realizing until the end.]

  28. https://www.gwern.net/Timing

  29. ⁠, Matt Lakeman (2019-04-29):

    I read on a whim when came out. I’ve since gone through the audiobook 3.5 times and can confidently say it’s one of my favorite books of all time. I expected just to hear funny anecdotes about the making of and ⁠, but I found so much more depth. In my eyes, Disaster Artist is an examination of insanity (which I am defining as “the inability to perceive reality to the degree of low or non-functionality in regular life”). The book is a pushback against a subtle cultural norm that sees crazy people as having some sort of gift or potential or insight that everyone else doesn’t.

    This message hit me especially hard because I had my first real experience with a crazy person only a few months before I read Disaster Artist…We fired our employee. We offered a small severance, about ¼ of his monthly salary, just to smooth things over. The employee demanded a full month’s salary, which he said he needed to provide for his wife and child. Then he (an ex-marine) threatened to personally kill me if we didn’t pay him.

    That 30 minute phone call was terrifying. I wasn’t actually scared of being murdered, and we never gave in to his demands, but it wasn’t until that call that I understood what it meant to be crazy. It unnerved me in a sort of staring into the abyss way. This man was truly detached from reality. He either didn’t know or could not understand the facts before him. When presented with reality, he would lash out in pain and anguish and fury at phantom targets. I would make calm, reasonable arguments about how he had violated his work contract, hurt our business, hurt our clients, and lied to us, and he would respond with nonsensical excuses, random tangents, blaming his personal life, and never ever coming close to acknowledging his own culpability.

    I came away from the conversation with a mixture of pity, revulsion, and dread. I don’t know if this guy was bipolar, drug-addled, ⁠, or what, but I was 100% sure that this man lived in a nightmare. Everything was confusing and nonsensical to him. The world was dark, malevolent, and couldn’t stop hurting him even as he tried his best. I had an imagine of him sitting alone in his tiny apartment listening to that one student’s song over-and-over again on repeat while his mind blurred between random scientific and historical topics until he could no longer fight the urge to pick up the phone and call me or someone like me who took enough pity on him to politely listen for a few minutes until we made excuses and left him back alone in silence.

    I see Tommy Wiseau, the creator of The Room and the subject of Disaster Artist, in the same category as my ex-employee. The form of their insanity is somewhat different, but both men live tortured, miserable lives, and constantly lash out at bystanders because of it. However, unlike my ex-employee, Wiseau is beloved by the masses precisely for his insanity. This is a dangerous, inaccurate, unfair reality, and in my opinion, is precisely what the Disaster Artist book argues against.

  30. ⁠, Matt Lakeman (2020-01-22):

    When I first finished ⁠, like so many other players, I was disappointed. MGSV was supposed to be the “Missing Link” in the canon. It was that game that would reveal the bridge between the heroic of ⁠, ⁠, and ⁠, and the grand historical villain of and ⁠. As expressed by numerous launch trailers and tweets, MGSV was going to be a tale of Big Boss’s fall into darkness, driven by an insatiable lust for revenge, a consummate anger lit by his enemies which would scorch his soul until nothing was left but a power-hungry mad man who would threaten the world with nuclear war for the sake of his deluded ambitions. Instead we got an incredibly weird twist which did little more than retcon patch a largely ignored plot hole in one of the least-played Metal Gear games. We found out that the final boss of Metal Gear 1 was not Big Boss, but a body double, who through surgery and hypnotherapy was made into almost an exact copy of the legendary soldier. Again, like most other players, when I first finished the game I thought this was a neat trick, a typically crazy, convoluted, but seductively entertaining twist from one of my favorite storytellers of all time. But of course… it was also a major let down.

    …It wasn’t until I had put over 200 hours into my save file and replayed the entire game for a second time that the impact of Metal Gear Solid V’s story really hit me. Not only does MGSV do exactly what it was advertised to do—reveal the descent of Big Boss from hero to villain—but it does so in a subtle and narratively ambitious manner at a depth not seen in any video game since ⁠.

    MGSV is the story of Big Boss’s fall from grace, but it’s also so much more than that. MGSV may very well be Kojima’s magnum opus. The game distills all of the Metal Gear series’ most important thematic elements into a relatively simple story with a deceptively small scale. The reason the vast majority of players didn’t realize this is because, well, Kojima can be too subtle for his own good…MGSV really is about Big Boss becoming a horrible monster worthy of every conceivable condemnation. But that story is the bedrock layer hidden beneath a million other narrative layers designed to confuse and manipulate the player, in exactly the same way Big Boss and Zero’s whole Phantom Snake project was designed to confuse and manipulate Venom Snake.

  31. #columns

  32. #unicode-characters

  33. #matt-lakeman

  34. #drop-caps

  35. #link-icons

  36. https://en.wikiquote.org/wiki/Tao_Te_Ching

  37. https://en.wiktionary.org/wiki/rimy

  38. https://hoogle.haskell.org/?hoogle=%28Monoid_a%29_%3D%3E_a_-%3E_a_-%3E_a

  39. https://hackage.haskell.org/package/Barracuda

  40. https://wiki.haskell.org/Tail_recursion

  41. ⁠, Noam Brown, Anton Bakhtin, Adam Lerer, Qucheng Gong (2020-07-27):

    The combination of deep and search at both training and test time is a powerful paradigm that has led to a number of successes in single-agent settings and perfect-information games, best exemplified by ⁠. However, prior algorithms of this form cannot cope with imperfect-information games.

    This paper presents ReBeL, a general framework for self-play reinforcement learning and search that provably converges to a Nash equilibrium in any two-player zero-sum game. In the simpler setting of perfect-information games, ReBeL reduces to an algorithm similar to ⁠.

    Results in two different imperfect-information games show ReBeL converges to an approximate Nash equilibrium. We also show ReBeL achieves superhuman performance in no-limit ⁠, while using far less domain knowledge than any poker AI.

  42. Faces

  43. 2020-east.pdf: ⁠, Patricia East, Jenalee Doom, Estela Blanco, Raquel Burrows, Betsy Lozoff, Sheila Gahagan (2020-08-11; iq):

    Objective: This study examined how the lower cognitive skills in children who consumed iron-fortified formula in infancy relate to outcomes in young adulthood.

    Methods: Participants were 443 Chilean young adults (Mage = 21.2y, 55% female) who took part in a randomized controlled iron-deficiency anemia preventive trial during infancy (6–12 m). Slightly over half of participants (n = 237) received iron-fortified formula (12.7 mg/​​​​L) and 206 received a low-iron formula (2.3 mg/​​​​L). Spatial memory, IQ, and visual-motor integration were measured at age 10, and neurocognition, emotion regulation, educational level, and attainment of adult developmental milestones were assessed at age 21.

    Results: Consumption of iron-fortified formula in infancy was associated with poorer performance on neurocognitive tests in childhood, and these effects related to poorer neurocognitive, emotional, and educational outcomes in young adulthood. Dosage effects associated with consumption of iron-fortified formula were found for lower educational attainment and, marginally, slower mental processing. Those who received iron-fortified formula and had low age 10 cognitive abilities performed most poorly on neurocognitive tests at age 21.

    Conclusion: Findings suggest that the long-term development of infants who consume iron-fortified formula may be adversely affected.

    Clinical Trials number: NCT01166451.

    [Keywords: Iron supplementation, neurocognition, emotion regulation, memory, Chile]

  44. https://www.youtube.com/watch?v=13CZPWmke6A

  45. Killing-Rabbits

  46. deepmind.svg

  47. https://academic.oup.com/ije/article/43/3/775/758445

  48. http://amazon.com/gp/product/B0050MYHBQ/

  49. https://amzn.com/

  50. https://statmodeling.stat.columbia.edu/2004/12/29/type_1_type_2_t/

  51. https://statmodeling.stat.columbia.edu/2013/12/17/replication-backlash/

  52. https://www.animenewsnetwork.com/anime-spotlight/2018/summer/revue-starlight/.132471

  53. https://arstechnica.com/apple/news/2011/09/the-post-jobs-era-tim-cook-brings-philanthropy-back-to-apple.ars

  54. ⁠, Mark Braverman, Elchanan Mossel (2007-07-06):

    In this paper we study noisy sorting without re-sampling. In this problem there is an unknown order aπ(1) < … < aπ(n) where π is a permutation on n elements. The input is the status of ([^*n*^~2~]{.supsub}) queries of the form q(ai, xj), where q(ai, aj) = + with probability at least 1⁄2 +γ if π(i) > π(j) for all pairs ij, where γ > 0 is a constant and q(ai, aj) = −q(aj, ai) for all i and j. It is assumed that the errors are independent. Given the status of the queries the goal is to find the order. In other words, the goal is find a permutation σ that minimizes the number of pairs σ(i) > σ(j) where q(σ(i), σ(j)) = −. The problem so defined is the problem on distributions of inputs, each of which is a tournament obtained as a noisy perturbations of a linear order. Note that when γ < 1⁄2 and n is large, it is impossible to recover the original order π.

    It is known that the weighted feedback arc set problem on tournaments is in general. Here we present an algorithm of running time n𝒪 (γ−4)> and sampling complexity Oγ(n log n) that with high probability solves the noisy sorting without re-sampling problem. We also show that if aσ(1), aσ(2), …, aσ(n) is an optimal solution of the problem then it is “close” to the original order. More formally, with high probability it holds that ∑i|σ(i) − π(i)| = Θ(n) and maxi|σ(i) − π(i)| = Θ(log n).

    Our results are of interest in applications to ranking, such as ranking in sports, or ranking of search items based on comparisons by experts.

  55. https://www.bbc.com/news/business-43365710

  56. http://news.bbc.co.uk/2/hi/8448731.stm

  57. ⁠, James E. DiCarlo, Alejandro Chavez, Sven L. Dietz, Kevin M. Esvelt, George M. Church (2015-03-19):

    Inheritance-biasing “gene drives” may be capable of spreading genomic alterations made in laboratory organisms through wild populations. We previously considered the potential for RNA-guided gene drives based on the versatile genome editing system to serve as a general method of altering populations1. Here we report molecularly contained gene drive constructs in the yeast Saccharomyces cerevisiae that are typically copied at rates above 99% when mated to wild yeast. We successfully targeted both non-essential and essential genes, showed that the inheritance of an unrelated “cargo” gene could be biased by an adjacent drive, and constructed a drive capable of overwriting and reversing changes made by a previous drive. Our results demonstrate that RNA-guided gene drives are capable of efficiently biasing inheritance when mated to wild-type organisms over successive generations.

  58. ⁠, Daniel F. Levey, Murray B. Stein, Frank R. Wendt, Gita A. Pathak, Hang Zhou, Mihaela Aslan, Rachel Quaden, Kelly M. Harrington, Gerard Sanacora, Andrew M. McIntosh, John Concato, Renato Polimanti, Joel Gelernter, on behalf of the Million Veteran Program (2020-05-22):

    We report a large of depression using data from the Million Veteran Program (MVP), 23andMe Inc., ⁠, and FinnGen; including individuals of European ancestry (n = 1,154,267; 340,591 cases) and African ancestry (n = 59,600; 25,843 cases). We identified 223 and 233 independent SNPs associated with depression in European ancestry and transancestral analysis, respectively. within the MVP cohort across diagnosis, survey self-report of diagnosis, and a 2-item depression screen exceeded 0.81. Using transcriptome-wide association study (TWAS) we found statistically-significant associations for gene expression in several brain regions, including hypothalamus (NEGR1, p = 3.19×10−25) and nucleus accumbens (DRD2, p = 1.87×10−20). 178 genomic risk loci were fine-mapped to find likely causal variants. We identified likely pathogenicity in these variants and overlapping gene expression for 17 genes from our TWAS, including TRAF3. This study sheds light on the genetic architecture of depression and provides new insight into the interrelatedness of complex psychiatric traits.

  59. http://en.bitcoin.it/wiki/Witcoin

  60. http://bitcointalk.org/index.php?topic=82952.0;all

  61. https://www.bloomberg.com/news/2011-03-31/why-unemployment-rose-so-much-dropped-so-fast-commentary-by-alan-krueger.html

  62. http://bjo.bmj.com/content/93/8/997

  63. https://deepmind.com/blog/article/alphastar-mastering-real-time-strategy-game-starcraft-ii

  64. 2016-silver.pdf#deepmind: ⁠, David Silver, Aja Huang, Chris J. Maddison, Arthur Guez, Laurent Sifre, George van den Driessche, Julian Schrittwieser, Ioannis Antonoglou, Veda Panneershelvam, Marc Lanctot, Sander Dieleman, Dominik Grewe, John Nham, Nal Kalchbrenner, Ilya Sutskever, Timothy Lillicrap, Madeleine Leach, Koray Kavukcuoglu, Thore Graepel, Demis Hassabis (2016-01-28; reinforcement-learning):

    The game of Go has long been viewed as the most challenging of classic games for artificial intelligence owing to its enormous search space and the difficulty of evaluating board positions and moves. Here we introduce a new approach to computer Go that uses ‘value networks’ to evaluate board positions and ‘policy networks’ to select moves. These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. Without any lookahead search, the neural networks play Go at the level of state-of-the-art programs that simulate thousands of random games of self-play. We also introduce a new search algorithm that combines with value and policy networks. Using this search algorithm, our program AlphaGo achieved a 99.8% winning rate against other Go programs, and defeated the human European Go champion by 5 games to 0. This is the first time that a computer program has defeated a human professional player in the full-sized game of Go, a feat previously thought to be at least a decade away.

    [Anecdote: I hear from Groq that the original AlphaGo GPU implementation was not on track to defeat Lee Sedol by about a month before, when they happened to gamble on implementing TPUv1 support. The additional compute led to drastic performance gains, and the TPU model could beat the model in ~98 of 100 games, and the final model solidly defeated Lee Sedol. (Since TPUv1s reportedly only did inferencing/​​​​forward-mode, presumably they were not used for the initial imitation learning, or the policy gradients self-play, but for generating the ~30 million self-play games which the value network was trained on (doing regression/​​​​prediction of ‘board → P(win)’, requiring no state or activations from the self-play games, just generating an extremely large corpus which could be easily used by GPU training.]

  65. ⁠, David Silver, Hado van Hasselt, Matteo Hessel, Tom Schaul, Arthur Guez, Tim Harley, Gabriel Dulac-Arnold, David Reichert, Neil Rabinowitz, Andre Barreto, Thomas Degris (2016-12-28):

    One of the key challenges of artificial intelligence is to learn models that are effective in the context of planning. In this document we introduce the architecture. The predictron consists of a fully abstract model, represented by a Markov reward process, that can be rolled forward multiple “imagined” planning steps. Each forward pass of the predictron accumulates internal rewards and values over multiple planning depths. The predictron is trained end-to-end so as to make these accumulated values accurately approximate the true value function. We applied the predictron to procedurally generated random mazes and a simulator for the game of pool. The predictron yielded statistically-significantly more accurate predictions than conventional deep neural network architectures.

  66. http://distill.pub/2016/augmented-rnns/

  67. http://dl.dropbox.com/u/85192141/2012-scannell.pdf

  68. http://dl.dropboxusercontent.com/u/182368464/umineko-compress.tar.xz

  69. https://www.economist.com/briefing/2017/02/18/hello-again-dolly

  70. https://www.erowid.org

  71. 2020-bell.pdf#facebook: ⁠, Sean Bell, Yiqun Liu, Sami Alsheikh, Yina Tang, Ed Pizzi, M. Henning, Karun Singh, Omkar Parkhi, Fedor Borisyuk (2020-08-22; ai):

    In this paper, we present GrokNet, a deployed image recognition system for commerce applications. GrokNet leverages a multi-task learning approach to train a single computer vision trunk. We achieve a 2.1× improvement in exact product match accuracy when compared to the previous state-of-the-art Facebook product recognition system. We achieve this by training on 7 datasets across several commerce verticals, using 80 categorical loss functions and 3 embedding losses. We share our experience of combining diverse sources with wide-ranging label semantics and image statistics, including learning from human annotations, user-generated tags, and noisy search engine interaction data. GrokNet has demonstrated gains in production applications and operates at Facebook scale.

  72. https://ai.facebook.com/blog/a-highly-efficient-real-time-text-to-speech-system-deployed-on-cpus/

  73. ⁠, Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston (2020-04-28):

    Building open-domain chatbots is a challenging area for machine learning research. While prior work has shown that scaling neural models in the number of parameters and the size of the data they are trained on gives improved results, we show that other ingredients are important for a high-performing chatbot. Good conversation requires a number of skills that an expert conversationalist blends in a seamless way: providing engaging talking points and listening to their partners, and displaying knowledge, empathy and personality appropriately, while maintaining a consistent persona. We show that large scale models can learn these skills when given appropriate training data and choice of generation strategy. We build variants of these recipes with 90M, 2.7B and 9.4B parameter models, and make our models and code publicly available. Human evaluations show our best models are superior to existing approaches in multi-turn dialogue in terms of engagingness and humanness measurements. We then discuss the limitations of this work by analyzing failure cases of our models.

  74. https://fis.fda.gov/sense/app/d10be6bb-494e-4cd2-82e4-0135608ddc13/sheet/45beeb74-30ab-46be-8267-5756582633b4/state/analysis

  75. https://www.filfre.net/2016/08/ibms-new-flavor/

  76. http://kpreid.github.com/mathquiz/mathquiz.html

  77. https://www.goodreads.com/api

  78. https://www.google.com/about/company/history/

  79. ⁠, Azalia Mirhoseini, Hieu Pham, Quoc V. Le, Benoit Steiner, Rasmus Larsen, Yuefeng Zhou, Naveen Kumar, Mohammad Norouzi, Samy Bengio, Jeff Dean (2017-06-13):

    The past few years have witnessed a growth in size and computational requirements for training and inference with neural networks. Currently, a common approach to address these requirements is to use a heterogeneous distributed environment with a mixture of hardware devices such as CPUs and GPUs. Importantly, the decision of placing parts of the neural models on devices is often made by human experts based on simple heuristics and intuitions. In this paper, we propose a method which learns to optimize device placement for TensorFlow computational graphs. Key to our method is the use of a sequence-to-sequence model to predict which subsets of operations in a TensorFlow graph should run on which of the available devices. The execution time of the predicted placements is then used as the reward signal to optimize the parameters of the sequence-to-sequence model. Our main result is that on Inception-V3 for classification, and on LSTM, for language modeling and neural machine translation, our model finds non-trivial device placements that outperform hand-crafted heuristics and traditional algorithmic methods.

  80. https://arxiv.org/pdf/2009.06732.pdf#org=google&page=6

  81. https://www.greaterwrong.com/posts/37sHjeisS9uJufi4u/scholarship-how-to-do-it-efficiently

  82. http://lesswrong.com

  83. https://www.harney.com/

  84. https://www.haskell.org/

  85. https://www.gwern.net/2014-spirulina

  86. http://intelligence.org/2013/10/03/proofs/

  87. https://wayback.archive-it.org/org-350/20180911191924/https://www.nlm.nih.gov/news/calhoun_papers_released.html

  88. http://blog.archive.org/2011/08/17/scanning-a-braille-playboy/

  89. http://archiveteam.org/index.php?title=Google_Reader

  90. http://timetravel.mementoweb.org/

  91. http://replay.waybackmachine.org/19981202185145/http://www.ex.org/2.4/11-news.html

  92. https://www.webcitation.org/6Qj7v6mqd

  93. https://www.justice.gov/archive/usao/cac/Pressroom/2012/045.html

  94. https://kk.org/books/out-of-control/

  95. https://link.springer.com/article/10.1007%2FBF02253535

  96. http://longbets.org/

  97. http://blog.longnow.org/02014/08/21/lenski-long-term-evolution-experiment/

  98. http://groups.google.com/group/ankisrs/

  99. http://groups.yahoo.com/group/givewell/message/287

  100. https://gwern.substack.com

  101. https://www.tinyletter.com/

  102. https://www.mail-archive.com/cryptography@metzdowd.com/msg09959.html

  103. https://marginalrevolution.com/

  104. http://cognitivemedium.com/srs-mathematics

  105. https://mega.nz/#!0JVxHQCD!C7ijBpRWNpcL_gubWFR-GTBDJTW1jXI6ThzSxwaw2aE

  106. ⁠, Kevin Simler (2019-05-13):

    [Interactive Javascript visualizations of epidemiology: how infection rates, immunity, reinfections, topology, and infection density all yield supercritical or subcritical explosions, with thought-example of science as a network community infected by careerism/​​​​Replication-Crisis problems.]

    If you’ve spent any time thinking about complex systems, you surely understand the importance of networks. Networks rule our world. From the chemical reaction pathways inside a cell, to the web of relationships in an ecosystem, to the trade and political networks that shape the course of history. Or consider this very post you’re reading. You probably found it on a social network, downloaded it from a computer network, and are currently deciphering it with your neural network.

    But as much as I’ve thought about networks over the years, I didn’t appreciate (until very recently) the importance of simple diffusion. This is our topic for today: the way things move and spread, somewhat chaotically, across a network. Some examples to whet the appetite:

    • Infectious diseases jumping from host to host within a population
    • Memes spreading across a follower graph on social media
    • A wildfire breaking out across a landscape
    • Ideas and practices diffusing through a culture
    • Neutrons cascading through a hunk of enriched uranium

    A quick note about form. Unlike all my previous work, this essay is interactive. There will be sliders to pull, buttons to push, and things that dance around on the screen. I’m pretty excited about this, and I hope you are too.

  107. https://mlp.fandom.com/wiki/A_Canterlot_Wedding_-_Part_1

  108. https://myanimelist.net/anime/1370/Atama_Yama

  109. http://blogs.nature.com/news/2011/09/reliability_of_new_drug_target.html

  110. https://www.newyorker.com/books/page-turner/the-mystery-of-s-the-man-with-an-impossible-memory

  111. http://17th-angel.tumblr.com/post/11409371268/anno-a-transfer-student-opens-the-door-with-a

  112. https://forum.evageeks.org/

  113. https://www.evamonkey.com/ask-john/has-evangelion-influenced-contemporary-gundam-anime.php

  114. http://evaotaku.com/html/programbooks.html

  115. http://gainax.co.jp/

  116. http://khara.co.jp/hideakianno/personal-biography.html

  117. http://neweva.blog103.fc2.com/blog-entry-1198.html

  118. https://eva.onegeek.org/

  119. ⁠, Park, Bum Jin Tsunetsugu, Yuko Kasetani, Tamami Kagawa, Takahide Miyazaki, Yoshifumi (2010):

    This paper reviews previous research on the physiological effects of Shinrin-yoku (taking in the forest atmosphere or forest bathing), and presents new results from field experiments conducted in 24 forests across Japan. The term Shinrin-yoku was coined by the Japanese Ministry of Agriculture, Forestry, and Fisheries in 1982, and can be defined as making contact with and taking in the atmosphere of the forest. In order to clarify the physiological effects of Shinrin-yoku, we conducted field experiments in 24 forests across Japan. In each experiment, 12 subjects (280 total; ages 21.7 ± 1.5 year) walked in and viewed a forest or city area. On the first day, six subjects were sent to a forest area, and the others to a city area. On the second day, each group was sent to the other area as a cross-check. Salivary cortisol, blood pressure, pulse rate, and heart rate variability were used as indices. These indices were measured in the morning at the accommodation facility before breakfast and also both before and after the walking (for 16 ± 5 min) and viewing (for 14 ± 2 min). The R-R interval was also measured during the walking and viewing periods. The results show that forest environments promote lower concentrations of cortisol, lower pulse rate, lower blood pressure, greater parasympathetic nerve activity, and lower sympathetic nerve activity than do city environments. These results will contribute to the development of a research field dedicated to forest medicine, which may be used as a strategy for preventive medicine.

  120. https://6thfloor.blogs.nytimes.com/2013/03/20/a-sham-procedure-leads-to-disappointing-m-s-news/

  121. https://online.wsj.com/article/SB10000872396390443696604577647870908169992.html

  122. 2020-chen.pdf#openai: ⁠, Mark Chen, Alec Radford, Rewon Child, Jeff Wu, Heewoo Jun, Prafulla Dhariwal, David Luan, Ilya Sutskever () (2020-06-17; ai):

    Inspired by progress in unsupervised representation learning for natural language, we examine whether similar models can learn useful representations for images. We train a sequence Transformer to auto-regressively predict pixels, without incorporating knowledge of the 2D input structure. Despite training on low-resolution ImageNet without labels, we find that a scale model learns strong image representations as measured by linear probing, fine-tuning, and low-data classification. On CIFAR-10, we achieve 96.3% accuracy with a linear probe, outperforming a supervised Wide ⁠, and 99.0% accuracy with full fine-tuning, matching the top supervised pre-trained models. An even larger model trained on a mixture of ImageNet and web images is competitive with self-supervised benchmarks on ImageNet, achieving 72.0% top-1 accuracy on a linear probe of our features.

    [See also ⁠.]

  123. ⁠, Alec Radford, Jeffrey Wu, Dario Amodei, Daniela Amodei, Jack Clark, Miles Brundage, Ilya Sutskever (OpenAI) (2019-02-14):

    Our model, called GPT-2 (a successor to GPT), was trained simply to predict the next word in 40GB of Internet text. Due to our concerns about malicious applications of the technology, we are not releasing the trained model. As an experiment in responsible disclosure, we are instead releasing a much smaller model for researchers to experiment with, as well as a ⁠.

    GPT-2 is a large -based language model with 1.5 billion parameters, trained on a dataset of 8 million web pages. GPT-2 is trained with a simple objective: predict the next word, given all of the previous words within some text. The diversity of the dataset causes this simple goal to contain naturally occurring demonstrations of many tasks across diverse domains. GPT-2 is a direct scale-up of GPT, with more than 10× the parameters and trained on more than 10× the amount of data.

    GPT-2 displays a broad set of capabilities, including the ability to generate conditional synthetic text samples of unprecedented quality, where we prime the model with an input and have it generate a lengthy continuation. In addition, GPT-2 outperforms other language models trained on specific domains (like Wikipedia, news, or books) without needing to use these domain-specific training datasets. On language tasks like question answering, reading comprehension, summarization, and translation, GPT-2 begins to learn these tasks from the raw text, using no task-specific training data. While scores on these downstream tasks are far from state-of-the-art, they suggest that the tasks can benefit from unsupervised techniques, given sufficient (unlabeled) data and compute.

  124. ⁠, Yan Duan, John Schulman, Xi Chen, Peter L. Bartlett, Ilya Sutskever, Pieter Abbeel (2016-11-09):

    Deep reinforcement learning (deep RL) has been successful in learning sophisticated behaviors automatically; however, the learning process requires a huge number of trials. In contrast, animals can learn new tasks in just a few trials, benefiting from their prior knowledge about the world. This paper seeks to bridge this gap. Rather than designing a “fast” reinforcement learning algorithm, we propose to represent it as a recurrent neural network (RNN) and learn it from data. In our proposed method, RL2, the algorithm is encoded in the weights of the RNN, which are learned slowly through a general-purpose (“slow”) RL algorithm. The RNN receives all information a typical RL algorithm would receive, including observations, actions, rewards, and termination flags; and it retains its state across episodes in a given (MDP). The activations of the RNN store the state of the “fast” RL algorithm on the current (previously unseen) MDP. We evaluate RL2 experimentally on both small-scale and large-scale problems. On the small-scale side, we train it to solve randomly generated multi-arm bandit problems and finite MDPs. After RL2 is trained, its performance on new MDPs is close to human-designed algorithms with optimality guarantees. On the large-scale side, we test RL2 on a vision-based navigation task and show that it scales up to high-dimensional problems.

  125. ⁠, Robin Hanson (2009-07-04):

    A better intuition for common abilities can be found by browsing the US National Assessment of Adult Literacy sample questions⁠.

    For example, in 1992 out of a random sample of US adults, 7% could not do item SCOR300, which is to find the expiration date on a driver’s license. 26% could not do item AB60303, which is to check the “Please Call” box on a phone message slip when they’ve been told:

    James Davidson phones and asks to speak with Ann Jones, who is at a meeting. He needs to know if the contracts he sent are satisfactory and requests that she call before 2:00PM. His number is 259-3860. Fill in the message slip below.

    Only 52% could do item AB30901, which is to look at a table on page 118 of the 1980 World Almanac and answer:

    According to the chart, did U.S. exports of oil (petroleum) increase or decrease between 1976 and 1978?

  126. https://www.patreon.com/gwern

  127. ⁠, Adam R. Boyko, Pascale Quignon, Lin Li, Jeffrey J. Schoenebeck, Jeremiah D. Degenhardt, Kirk E. Lohmueller, Keyan Zhao, Abra Brisbin, Heidi G. Parker, Bridgett M. vonHoldt, Michele Cargill, Adam Auton, Andy Reynolds, Abdel G. Elkahloun, Marta Castelhano, Dana S. Mosher, Nathan B. Sutter, Gary S. Johnson, John Novembre, Melissa J. Hubisz, Adam Siepel, Robert K. Wayne, Carlos D. Bustamante, Elaine A. Ostrander (2010-07-02):

    Domestic dogs exhibit tremendous phenotypic diversity, including a greater variation in body size than any other terrestrial mammal. Here, we generate a high density map of canine genetic variation by genotyping 915 dogs from 80 domestic dog breeds, 83 wild canids, and 10 outbred African shelter dogs across 60,968 single-nucleotide polymorphisms (SNPs).

    Coupling this genomic resource with external measurements from breed standards and individuals as well as skeletal measurements from museum specimens, we identify 51 regions of the dog genome associated with phenotypic variation among breeds in 57 traits. The complex traits include average breed body size and external body dimensions and cranial, dental, and long bone shape and size with and without ⁠. In contrast to the results from association mapping of quantitative traits in humans and domesticated plants, we find that across dog breeds, a small number of (≤3) explain the majority of phenotypic variation for most of the traits we studied. In addition, many genomic regions show signatures of recent selection, with most of the highly differentiated regions being associated with breed-defining traits such as body size, coat characteristics, and ear floppiness.

    Our results demonstrate the efficacy of mapping multiple traits in the domestic dog using a database of genotyped individuals and highlight the important role human-directed selection has played in altering the genetic architecture of key traits in this important species.

    Author Summary: Dogs offer a unique system for the study of genes controlling morphology. DNA from 915 dogs from 80 domestic breeds, as well as a set of feral dogs, was tested at over 60,000 points of variation and the dataset analyzed using novel methods to find loci regulating body size, head shape, leg length, ear position, and a host of other traits. Because each dog breed has undergone strong selection by breeders to have a particular appearance, there is a strong footprint of selection in regions of the genome that are important for controlling traits that define each breed. These analyses identified new regions of the genome, or loci, that are important in controlling body size and shape. Our results, which feature the largest number of domestic dogs studied at such a high level of genetic detail, demonstrate the power of the dog as a model for finding genes that control the body plan of mammals. Further, we show that the remarkable diversity of form in the dog, in contrast to some other species studied to date, appears to have a simple genetic basis dominated by genes of major effect.

  128. https://speakingofmedicine.plos.org/2012/06/25/less-research-is-needed/

  129. http://www.plosmedicine.org/article/fetchObject.action?uri=info:doi/10.1371/journal.pmed.0020124.t004&representation=PNG_M

  130. ⁠, Hikaru Takeuchi, Yasuyuki Taki, Yuko Sassa, Hiroshi Hashizume, Atsushi Sekiguchi, Ai Fukushima, Ryuta Kawashima (2011-07-11):

    Training (WM) improves performance on untrained cognitive tasks and alters functional activity. However, WM training’s effects on gray matter morphology and a wide range of cognitive tasks are still unknown. We investigated this issue using voxel-based morphometry (VBM), various psychological measures, such as non-trained WM tasks and a creativity task, and intensive adaptive training of WM using mental calculations (IATWMMC), all of which are typical WM tasks. IATWMMC was associated with reduced regional gray matter volume in the bilateral fronto-parietal regions and the left superior temporal gyrus. It improved verbal letter span and complex arithmetic ability, but deteriorated creativity. These results confirm the training-induced plasticity in psychological mechanisms and the plasticity of gray matter structures in regions that have been assumed to be under strong genetic control.

  131. ⁠, Jane Bradbury (2005-03-15):

    As a species, we pride ourselves on the uniqueness of our brain. Relative to our body size, the human brain is bigger than that of any other animal. It may also contain unique structures and patterns of organisation that presumably underlie our intelligence and ability to manipulate our environment. But how did our unique brain originate, and under what selective pressures did it evolve? Some of the answers may lie in the genetic differences that researchers are now uncovering between us and our closest relatives.

    ToC: Costs and Benefits · The Genetics of Human Brain Evolution · Enter · Gene Expression · Scratching at the Surface · Further Reading

  132. ⁠, Richard J. Hatchett, Carter E. Mecher, Marc Lipsitch (2007-05-01):

    Nonpharmaceutical interventions (NPIs) intended to reduce infectious contacts between persons form an integral part of plans to mitigate the impact of the next influenza pandemic. Although the potential benefits of NPIs are supported by mathematical models, the historical evidence for the impact of such interventions in past pandemics has not been systematically examined. We obtained data on the timing of 19 classes of NPI in 17 U.S. cities during the 1918 pandemic and tested the hypothesis that early implementation of multiple interventions was associated with reduced disease transmission. Consistent with this hypothesis, cities in which multiple interventions were implemented at an early phase of the epidemic had peak death rates ≈50% lower than those that did not and had less-steep epidemic curves. Cities in which multiple interventions were implemented at an early phase of the epidemic also showed a trend toward lower cumulative excess mortality, but the difference was smaller (≈20%) and less than that for peak death rates. This finding was not unexpected, given that few cities maintained NPIs longer than 6 weeks in 1918. Early implementation of certain interventions, including closure of schools, churches, and theaters, was associated with lower peak death rates, but no single intervention showed an association with improved aggregate outcomes for the 1918 phase of the pandemic. These findings support the hypothesis that rapid implementation of multiple NPIs can statistically-significantly reduce influenza transmission, but that viral spread will be renewed upon relaxation of such measures.

    …In comparisons across cities (Figure 2a, Table 2), we found that aggressive early intervention was statistically-significantly associated with a lower peak of excess mortality (Spearman ρ = −0.49 to −0.68, p = 0.002–0.047; see Table 2, Number of interventions before, for the number of NPIs before a given CEPID cutoff vs. peak mortality). Cities that implemented three or fewer NPIs before 20/​​​​100,000 CEPID had a median peak weekly death rate of 146/​​​​100,000, compared with 65/​​​​100,000 in those implementing four or more NPIs by that time (Figure 2a, p = 0.005). The relationship was similar for normalized peak death rates and for a range of possible cutoffs (see Table 2, CEPID at time of intervention), although the relationship became weaker as later interventions were included. Cities with more early NPIs also had fewer total excess deaths during the study period (Figure 2b, Table 2, 1918 total), but this association was weaker: cities with three or fewer NPIs before CEPID = 20/​​​​100,000 experienced a median total excess death rate of 551/​​​​100,000, compared with a median rate of 405/​​​​100,000 in cities with four or more NPIs (p = 0.03).

  133. https://predictionbook.com

  134. http://publicdomainreview.org/2013/08/07/the-lost-world-of-the-london-coffeehouse/

  135. http://cran.r-project.org/package=censReg

  136. ⁠, Benjamin Bahney, Howard J. Shatz, Carroll Ganier, Renny McPherson, Barbara Sude, Sara Beth Elson, Ghassan Schbley () (2010):

    This monograph analyzes the finances of the militant group al-Qa’ida in Iraq (AQI) in Anbar province during 2005 and 2006, at the peak of the group’s power and influence. The authors draw on captured documents that give details on the daily financial transactions of one specific sector within Anbar province and of the financial transactions of the AQI provincial administration. Some of their conclusions are: AQI was a hierarchical organization with decentralized decision-making; AQI in Anbar was profitable enough to send substantial revenues out of the province in 2006; AQI relied on extortion, theft, and black market sales to fund its operations in Anbar; AQI needed large, regular revenue sources to fund its operations, but its administrative leaders did not hold much cash on hand. The authors’ interpretation of data on compensation practices and participants’ risk of death indicates that AQI members were poorly compensated and suggests that they were not motivated primarily by money to join the group. The authors also find that mounting attacks required organizational expenditures well beyond the cost of material used in attacks. One major conclusion is that disrupting AQI’s financial flows could disrupt the pace of their attacks.

  137. http://en.reddit.com/r/Supplements/comments/mr0h1/taking_melatonin_forever/

  138. http://cro.sagepub.com/content/15/5/252.full.pdf+html

  139. http://www.salon.com:80/news/opinion/glenn_greenwald/2010/05/25/whistleblowers

  140. http://scholar.google.com/citations?user=9hEhCHYAAAAJ&hl=en&oi=ao

  141. http://scholars-stage.org/2014/04/meditations-on-maoism-ye-fus-hard-road.html

  142. https://www.sciencedirect.com/science/article/pii/S0002929717301076

  143. https://advances.sciencemag.org/content/4/9/eaar3620

  144. https://slate.com/cover-stories/2017/05/daryl-bem-proved-esp-is-real-showed-science-is-broken.html

  145. https://slatestarcodex.com/2015/01/15/depression-is-not-a-proxy-for-social-dysfunction/

  146. ⁠, Scott Alexander (2015-12-28):

    [Unsong is a finished (2015–2017) online web serial fantasy “kabbalah-punk” novel written by Scott Alexander (SSC). summary:

    Aaron Smith-Teller works in a kabbalistic sweatshop in Silicon Valley, where he and hundreds of other minimum-wage workers try to brute-force the Holy Names of God. All around him, vast forces have been moving their pieces into place for the final confrontation. An overworked archangel tries to debug the laws of physics. Henry Kissinger transforms the ancient conflict between Heaven and Hell into a US-Soviet proxy war. A Mexican hedge wizard with no actual magic wreaks havoc using the dark art of placebomancy. The Messiah reads a book by and starts wondering exactly what it would mean to do as much good as possible…

    Aaron doesn’t care about any of this. He and his not-quite-girlfriend Ana are engaged in something far more important—griping about magical intellectual property law. But when a chance discovery brings them into conflict with mysterious international magic-intellectual-property watchdog UNSONG, they find themselves caught in a web of plots, crusades, and prophecies leading inexorably to the end of the world.

    TVTropes⁠; my review of Unsong: ★★★★☆.]

  147. 2011-yvain-iliadaslawsuit.html

  148. https://stackoverflow.com/questions/1197575/can-scripts-be-inserted-with-innerhtml

  149. http://mathoverflow.net/questions/32967/have-any-long-suspected-irrational-numbers-turned-out-to-be-rational

  150. http://crypto.stackexchange.com/questions/2507/can-i-encrypt-user-input-in-a-way-i-cant-decrypt-it-for-a-certain-period-of-tim

  151. https://www.technologyreview.com/2011/06/21/193829/the-measured-life/

  152. https://www.theatlantic.com/business/archive/2011/06/beware-the-stunning-pilot-program/240352/

  153. https://www.theguardian.com/books/2013/jul/10/man-behind-dickens-dostoevsky-hoax

  154. https://www.theparisreview.org/blog/2018/04/25/the-strange-history-of-the-king-pine/

  155. https://econlolcats.tumblr.com/

  156. ⁠, TVTropes (2019):

    is the first full-length Gundam animated movie released in 1988. Char’s Counterattack is the culmination of the original saga begun in and continued through and ⁠, marking the final conflict of the fourteen-year rivalry between and ⁠, and the end of the Earth Federation/​​​​Zeon conflicts.

    …The movie is noteworthy for having a rather unusual genesis. Originally, director was going to wrap up Amuro and Char’s storyline in Gundam ZZ, but mid-way through production he was given the go-ahead to make a movie, forcing the plot of ZZ to be rewritten (details on its trope page). In the meantime Tomino wrote the novel Hi-Streamer, but when Sunrise gave him the green light, he went back and wrote a second novel, Beltorchika’s Children, which he specifically wrote to be adapted into a movie. However, Sunrise instead chose to use Hi-Streamer, with the final film being a pretty straightforward adaptation of its second half.

  157. https://developer.twitter.com/en/docs/twitter-api/v1/rules-and-filtering/search-operators

  158. https://uptontea.com/shopcart/item.asp?itemID=TT15

  159. http://lists.urth.net/pipermail/urth-urth.net/2010-December/019108.html

  160. https://www.washingtonpost.com/graphics/2018/investigations/dog-auction-rescue-groups-donations/

  161. http://blog.wikimedia.org/2009/11/26/wikipedias-volunteer-story/

  162. https://wikimediafoundation.org/wiki/Privacy_policy

  163. https://en.wikisource.org/wiki/Essays_on_Political_Economy/That_Which_Is_Seen,_and_That_Which_Is_Not_Seen

  164. https://en.wiktionary.org/wiki/bien_pensant

  165. http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD002966.pub3/epdf/standard

  166. http://archive.wired.com/geekdad/2012/01/everything-about-learning/

  167. https://www.cdc.gov/nchs/nvss/births.htm

  168. https://www.cell.com/ajhg/fulltext/S0002-9297(18)30405-1

  169. https://www.edge.org/conversation/alex_sandy_pentland-the-human-strategy

  170. https://www.fanfiction.net/s/10360716/1/The-Metropolitan-Man

  171. https://www.motherjones.com/kevin-drum/2018/02/an-updated-lead-crime-roundup-for-2018/

  172. https://www.nber.org/papers/w16082

  173. https://www.vanityfair.com/politics/2012/10/michael-lewis-profile-barack-obama

  174. ⁠, Ezra Klein (Vox) (2015-05-27):

    But lately, Gates has been obsessing over a dark question: what’s likeliest to kill more than 10 million human beings in the next 20 years? He ticks off the disaster movie stuff—“big volcanic explosion, gigantic earthquake, asteroid”—but says the more he learns about them, the more he realizes the probability is “very low.” Then there’s war, of course. But Gates isn’t that worried about war because the entire human race worries about war pretty much all the time, and the most dangerous kind of war, nuclear war, seems pretty contained, at least for now.

    But there’s something out there that’s as bad as war, something that kills as many people as war, and Gates doesn’t think we’re ready for it. “Look at the death chart of the 20th century”, he says, because he’s the kind of guy that looks at death charts. “I think everybody would say there must be a spike for World War I. Sure enough, there it is, like 25 million. And there must be a big spike for World War II, and there it is, it’s like 65 million. But then you’ll see this other spike that is as large as World War II right after World War I, and most people, would say, ‘What was that?’” “Well, that was the Spanish flu.”

    No one can say we weren’t warned. And warned. And warned. A pandemic disease is the most predictable catastrophe in the history of the human race, if only because it has happened to the human race so many, many times before…“You can’t use the word lucky or fortunate about something like Ebola that killed 10,000 people”, Klain says. “But it was the most favorable scenario for the world to face one of these things. Ebola is very difficult to transmit. Everyone who is contagious has a visible symptom. It broke out in three relatively small countries that don’t send many travelers to the US. And those three countries have good relationships with America and were welcoming of Western aid.” “With a pandemic flu, the disease would be much more contagious than Ebola”, Klain continues. “The people who are contagious may not have visible symptoms. It could break out in a highly populous country that sends thousands of travelers a day to the US. It could be a country with megacities with tens of millions of people. And it could be a country where sending in the 101st Airborne isn’t possible.”

    …Behind Gates’s fear of pandemic disease is an algorithmic model of how disease moves through the modern world. He funded that model to help with his foundation’s work eradicating polio. But then he used it to look into how a disease that acted like the Spanish flu of 1918 would work in today’s world. The results were shocking, even to Gates. “Within 60 days it’s basically in all urban centers around the entire globe”, he says. “That didn’t happen with the Spanish flu.”

  175. https://www.wsj.com/news/articles/SB10001424052702303380004579521482247869874

  176. http://paulgraham.com/hundred.html

  177. https://news.ycombinator.com/item?id=10012625

  178. http://aino.bandcamp.com/track/--2

  179. https://poniesatdawn.bandcamp.com/

  180. https://soundcloud.com/leggysalad/girls-afternoon-appointments#play

  181. https://www.youtube.com/watch?v=cG7v9eCq2u4&feature=youtu.be&t=33m49s

  182. https://www.youtube.com/channel/UCeNwyKuv5SMnN6ovlpbz1SQ

  183. https://yunnansourcing.com/

  184. https://what-if.xkcd.com/145/

  185. 2017-07-30-gs-pilot.csv

  186. 2014-tenijenhuis-supplement.doc

  187. 2015-mosing-supplement.docx

  188. 2012-zhong.ebt

  189. 1992-dreyfus-whatcomputerstillcantdo.epub

  190. 2013-c84-downloads.json

  191. 2020-06-08-danbooru2019-palm-handannotations-export.jsonl

  192. 2013-06-08-acircle-tohoarrange.mdb

  193. 2009-08-20-b3ta-fujitsuhtml.mht

  194. 2019-01-21-eric-socksurvey.ods

  195. rss-subscriptions.opml

  196. Lorem.page

  197. 2019-10-17-117m-poetry-cleanprojectgutenberg-samples.txt

  198. DeutscheZierschrift-M.ttf

  199. 2009-ling-data.xls

  200. 2010-nordhaus-nordhaus2007twocenturiesofproductivitygrowthincomputing-appendix.xlsx

  201. google-cse.xml

  202. ⁠, Gwern Branwen (2020-02-09):

    [Public-editable Google Docs document for coordinating a read through a large sample of neural-net-generated poetry, to locate the best poem samples for displaying in the GPT-2 writeup⁠.]

    I used a large neural net model, -1.5b, trained on hundreds of megabytes of poetry, to generate 1 million words of poetry. That’s too much for me to read by myself to find the best poems. Perhaps you’d like to help?

    It’s simple:

    1. Pick an unread URL from ‘Open Samples’ below, open it, and remove it from the list.
    2. Read it. (Each URL is ≤ 1000 lines, so it should be fun.)
    3. Add any good poems to ‘Selected Samples’ at the end of this document.
    4. Enjoy reading the current ‘Selected Samples’—or pick another URL to read!
  203. 1986-michie-onmachineintelligence.pdf#page=99: “On Machine Intelligence, Second Edition”⁠, Donald Michie

  204. 1962-bryson.pdf: ⁠, A. E. Bryson, W. F. Denham (1962-06-01; ai):

    A systematic and rapid steepest-ascent numerical procedure is described for solving two-point boundary-value problems in the calculus of variations for systems governed by a set of nonlinear ordinary differential equations. Numerical examples are presented for minimum time-to-climb and maximum altitude paths for a supersonic interceptor and maximum-range paths for an orbital glider.

    [Keywords: boundary-value problems, computer programming, differential equations, variational techniques]

    …A systematic and rapid steepest-ascent numerical procedure is described for determining optimum programs for nonlinear systems with terminal constraints. The procedure uses the concept of local linearization around a nominal (non-optimum) path. The effect on the terminal conditions of a small change in the control variable program is determined by numerical integration of the adjoint differential equations for small perturbations about the nominal path. Having these adjoint (or influence) functions, it is then possible to determine the change in the control variable program that gives maximum increase in the pay-off function for a given mean-square perturbation of the control variable program while simultaneously changing the terminal quantities by desired amounts. By repeating this process in small steps, a control variable program that minimizes one quantity and yields specified values of other terminal quantities can be approached as closely as desired. Three numerical examples are presented: (a) The angle-of-attack program for a typical supersonic interceptor to climb to altitude in minimum time is determined with and without specified terminal velocity and heading. (b) The angle-of-attack program for the same interceptor to climb to maximum altitude is determined, (c) The angle-of-attack program is determined for a hypersonic orbital glider to obtain maximum surface range starting from satellite speed at 300,000 ft altitude.

  205. http://citeseerx.ist.psu.edu/viewdoc/download?doi=

  206. http://eprint.iacr.org/2013/782

  207. https://pdfs.semanticscholar.org/00d3/6b267777b670abd1a3b98a21bf662245a7c4.pdf

  208. ⁠, Michael T. Wong, Douglas E. Zongker, David H. Salesin (1998):

    This paper describes some of the principles of traditional floral ornamental design, and explores ways in which these designs can be created algorithmically. It introduces the idea of “adaptive clip art”, which encapsulates the rules for creating a specific ornamental pattern. Adaptive clip art can be used to generate patterns that are tailored to fit a particularly shaped region of the plane. If the region is resized or reshaped, the ornament can be automatically regenerated to fill this new area in an appropriate way. Our ornamental patterns are created in two steps: first, the geometry of the pattern is generated as a set of two-dimensional curves and filled boundaries; second, this geometry is rendered in any number of styles. We demonstrate our approach with a variety of floral ornamental designs.

  209. morris.bmp

  210. 55403_m.gif

  211. favicon.ico

  212. 2010-1000enpark-tokyo-oota-heiwajimakoen.jpg

  213. 2011-gensowski-figure7-totaleffectofiqandpersonalityonlifetimeearnings.png

  214. video.svg

  215. businesscard-front-draft.xcf

  216. http://i.imgur.com/3Jb0b.jpg

  217. linkAbstract.R

  218. links.css

  219. hakyll.hs

  220. default.html

  221. 2019-12-22-gpt2-preferencelearning-gwern-abcmusic.patch

  222. markdown-lint.sh

  223. build_css.php

  224. http://www.collisiondetection.net/mt/archives/2013/07/wired_love_a_ta.php

  225. twdne.conf

  226. firmware-v2.6.3R-zeo.img

  227. https://hivemind-repo.s3-us-west-2.amazonaws.com/twdne3/twdne3.onnx

  228. michaellee-memoryretentionexperiments-data.tar

  229. 2015-06-03-karpathy-charrnn-visualization.tar.xz

  230. 2019-02-10-stylegan-holo-handselectedsamples.zip

  231. 2017-reddit-dhieno-theplace-timelapseevolution.mp4

  232. http://iqtest.dk/main.swf

  233. tea-mineralwaters-bestarm-sequential.webm

  234. 2020-03-06-fifteenai-fluttershy-sithcode.mp3

  235. batmantv.rm

  236. letterman_any_sense.wav

  237. https://1dollarscan.com/

  238. https://1d4chan.org/wiki/Tale_of_an_Industrious_Rogue,_Part_I

  239. http://80000hours.org/blog/170-estimation-part-i-how-to-do-it

  240. https://abandonedfootnotes.blogspot.com/2011/03/simple-model-of-cults-of-personality.html

  241. http://www.temcauley.staff.shef.ac.uk/waka0088.shtml

  242. https://www.academia.edu/24382120

  243. https://academictorrents.com/details/a306397ccf9c2ead27155983c254227c0fd938e2

  244. https://accidentallyquadratic.tumblr.com/

  245. ⁠, Sagar Savla (2019-02-04):

    The World Health Organization (WHO) estimates that there are 466 million people globally that are deaf and hard of hearing. A crucial technology in empowering communication and inclusive access to the world’s information to this population is automatic speech recognition (ASR), which enables computers to detect audible languages and transcribe them into text for reading. Google’s ASR is behind automated captions in Youtube, presentations in Slides and also phone calls…Today, we’re announcing Live Transcribe, a free Android service that makes real-world conversations more accessible by bringing the power of automatic captioning into everyday, conversational use. Powered by Google Cloud, Live Transcribe captions conversations in real-time, supporting over 70 languages and more than 80% of the world’s population. You can launch it with a single tap from within any app, directly from the accessibility icon on the system tray.

    …Relying on cloud ASR provides us greater accuracy, but we wanted to reduce the network data consumption that Live Transcribe requires. To do this, we implemented an on-device neural network-based speech detector, built on our previous work with AudioSet. This network is an image-like model, similar to our published VGGish model, which detects speech and automatically manages network connections to the cloud ASR engine, minimizing data usage over long periods of use.

    …Known as the cocktail party problem, understanding a speaker in a noisy room is a major challenge for computers. To address this, we built an indicator that visualizes the volume of user speech relative to background noise. This also gives users instant feedback on how well the microphone is receiving the incoming speech from the speaker, allowing them to adjust the placement of the phone…Potential future improvements in mobile-based automatic speech transcription include on-device recognition, speaker-separation, and speech enhancement. Relying solely on transcription can have pitfalls that can lead to miscommunication. Our research with Gallaudet University shows that combining it with other auditory signals like speech detection and a loudness indicator, makes a tangibly meaningful change in communication options for our users.

  246. ⁠, Nick Walton (2019-12):

    [AI Dungeon 2 is a project which trains -1.5b on logs from text adventure games; when used interactively by a human, it “plays RPG games with you, but because it is powered by GPT-2-1.5b, it is immensely flexible and can cope (to some degree) with almost any input, producing bizarre, hilarious, or surprisingly logical sequences of adventures. It became popular overnight, crushing Walton with bandwidth bills, and has been turned into an app and community to support distribution and development. See also https:/​​​​/​​​​colab.research.google.com/​​​​github/​​​​nickwalton/​​​​AIDungeon/​​​​blob/​​​​master/​​​​AIDungeon_2.ipynb and https:/​​​​/​​​​old.reddit.com/​​​​r/​​​​AIDungeon/ and https:/​​​​/​​​​web.archive.org/​​​​web/​​​​20191127163535/​​​​http:/​​​​/​​​​www.aidungeon.io/​​​​2019/​​​​11/​​​​my-orc-band-and-our-quest-for-equal.html .]

  247. http://www.ajcn.org/content/87/6/1715.long

  248. http://aleph.se/andart2/megascale/what-is-the-largest-possible-inhabitable-world/

  249. https://hn.algolia.com/?query=%22Candy%20Japan%22&sort=byPopularity&prefix&page=0&dateRange=all&type=all

  250. http://archpsyc.ama-assn.org/cgi/content/abstract/67/2/146

  251. http://www.amsciepub.com/doi/abs/10.2466/pr0.1981.49.2.470

  252. https://statmodeling.stat.columbia.edu/wp-content/uploads/2013/03/1154-Hauer-The-harm-done-by-tests-of-significance.pdf

  253. http://anidb.net/perl-bin/animedb.pl?show=lexicon&mode=character&vtype=ctag&relid=2296

  254. https://www.animenewsnetwork.com/brain-diving/2011-07-26

  255. http://www.alexa.com/siteinfo/ankisrs.net#

  256. https://apps.ankiweb.net/docs/manual.html#_javascript

  257. ⁠, Dena Zeraatkar, Mi Ah Han, Gordon H. Guyatt, Robin W. M. Vernooij, Regina El Dib, Kevin Cheung, Kirolos Milio, Max Zworth, Jessica J. Bartoszko, Claudia Valli, Montserrat Rabassa, Yung Lee, Joanna Zajac, Anna Prokop-Dorner, Calvin Lo, Malgorzata M. Bala, Pablo Alonso-Coello, Steven E. Hanna, Bradley C. Johnston (2019-11-19):

    Background: Dietary guidelines generally recommend limiting intake of red and processed meat. However, the quality of evidence implicating red and processed meat in adverse health outcomes remains unclear.

    Purpose: To evaluate the association between red and processed meat consumption and all-cause mortality, cardiometabolic outcomes, quality of life, and satisfaction with diet among adults.

    Data Sources: EMBASE (Elsevier), Cochrane Central Register of Controlled Trials (Wiley), Web of Science (Clarivate Analytics), CINAHL (EBSCO), and ProQuest from inception until July 2018 and MEDLINE from inception until April 2019, without language restrictions, as well as bibliographies of relevant articles.

    Study Selection: Cohort studies with at least 1000 participants that reported an association between unprocessed red or processed meat intake and outcomes of interest.

    Data Extraction: Teams of 2 reviewers independently extracted data and assessed risk of bias. One investigator assessed certainty of evidence, and the senior investigator confirmed the assessments.

    Data Synthesis: Of 61 articles reporting on 55 cohorts with more than 4 million participants, none addressed quality of life or satisfaction with diet. Low-certainty evidence was found that a reduction in unprocessed red meat intake of 3 servings per week is associated with a very small reduction in risk for cardiovascular mortality, stroke, myocardial infarction (MI), and type 2 diabetes. Likewise, low-certainty evidence was found that a reduction in processed meat intake of 3 servings per week is associated with a very small decrease in risk for all-cause mortality, cardiovascular mortality, stroke, MI, and type 2 diabetes.

    Limitation: Inadequate adjustment for known confounders, residual confounding due to observational design, and recall bias associated with dietary measurement.

    Conclusion: The magnitude of association between red and processed meat consumption and all-cause mortality and adverse cardiometabolic outcomes is very small, and the evidence is of low certainty.

  258. http://antilop.cc/sr/#the_chinese_connection

  259. http://www.antipope.org/charlie/fiction/coffee.html

  260. https://apenwarr.ca/log/?m=201707#04

  261. ⁠, Allen Institute For Artificial Intelligence (2019-02-26):

    This demonstration uses the public 345M 117M parameter OpenAI language model to generate sentences.

    Enter some initial text and the model will generate the most likely next words. You can click on one of those words to choose it and continue or just keep typing. Click the left arrow at the bottom to undo your last choice.

  262. http://archive.foolz.us/a/thread/77196171/#77207238

  263. http://archive.is/

  264. http://web.archive.org/web/20120628120817/http://blog.myzeo.com/sleep-forgetting-to-remember-to-forget/

  265. https://archive.today/u6ji#selection-236.0-236.1

  266. https://archivebox.io/

  267. http://www.archiveteam.org/index.php?title=ArchiveTeam_Warrior

  268. https://aur.archlinux.org/packages/mnemosyne/

  269. https://arstechnica.com/tech-policy/2015/01/silk-road-trial-federal-agent-explains-how-he-trapped-ulbricht/

  270. ⁠, Joel Simon (2019-09-09):

    [Artbreeder is an interactive generator website. Originally named “Ganbreeder” and providing only the 256px generator, it now provides a variety of BigGAN & StyleGAN models, including the anime portrait StyleGAN model. (It is more general than the similar Waifu Labs, but my anime model is not as good.) Users can generate random samples and explore slight variants of them to gradually explore the “latent space” and find interesting images, but they can also edit images more directly, upload existing images to find the most similar image produced by the model, etc. A popular website, it has generated >56m images from September 2019 to January 2020.]

  271. https://www.arxiv-vanity.com

  272. ⁠, Théophane Weber, Sébastien Racanière, David P. Reichert, Lars Buesing, Arthur Guez, Danilo Jimenez Rezende, Adria Puigdomènech Badia, Oriol Vinyals, Nicolas Heess, Yujia Li, Razvan Pascanu, Peter Battaglia, ⁠, David Silver, Daan Wierstra (2017-07-19):

    We introduce Imagination-Augmented Agents (I2As), a novel architecture for deep reinforcement learning combining model-free and model-based aspects. In contrast to most existing model-based reinforcement learning and planning methods, which prescribe how a model should be used to arrive at a policy, I2As learn to interpret predictions from a learned environment model to construct implicit plans in arbitrary ways, by using the predictions as additional context in deep policy networks. I2As show improved data efficiency, performance, and robustness to model misspecification compared to several baselines.

  273. http://ascii.textfiles.com/archives/3974

  274. ⁠, Luke Fater (Atlas Obscura) (2020-04-04):

    Like many of us, Seamus Blackley tweeted a photo of homemade sourdough while stuck at home this weekend. Unlike many of us, however, Blackley is no novice, and this was no online recipe comprised of everyday ingredients. A trained physicist and video game producer credited with inventing the Xbox, Blackley is also an experienced baker and amateur Egyptologist. The recipe came, in part, from ancient hieroglyphs, and the ingredients came, in part, from museum archives. Blackley’s weekend sourdough was the culmination of a year-long passion project that produced a loaf of bread not eaten for millennia. By extracting 4,500-year-old dormant yeast samples from ancient Egyptian baking vessels and reviving them in his home kitchen, Blackley and his collaborators quite literally brought history to life, and ate it.

    …Finally, Boston’s Museum of Fine Arts and Harvard’s Peabody Museum relented, allowing Blackley access to their deep troves of ancient Egyptian artifacts. “It was a little intimidating, to be honest”, he says. Bowman’s non-invasive extraction method resembles miniaturized fracking, wherein a portion of ceramic is injected with a nutrient bath before being pulled out through a syringe with the ancient yeast intact.

    Blackley shipped the samples to Bowman in Iowa, but took one vial home to Southern California to feed and propagate himself. Interestingly, the sample could only be revived with flour, a denser varietal of flour likely used by Egyptians of the Old Kingdom…Blackley revived the yeast and baked it on a pan in his conventional home oven, resulting in a loaf that made headlines this past August. “I’ve made a fuck ton of sourdough”, says Blackley, “but this was different.” The ancient loaves were sweeter and chewier than the standard modern sourdough, with a smooth crumb closer to white bread.

    …Her work showed that Egyptians placed their dough into a heated, conical, clay pot called a bedja before burying it in a hole surrounded by hot embers, a process Blackley made it his mission to reenact to perfection…He estimates that he cooked about 75 loaves before building his own bedja by hand and digging a hole in his backyard to master underground baking…From the ancient yeast to the underground baking method, Blackley had at long last produced an indisputably ancient loaf with the same rich sweetness as his summer sourdough.

  275. http://aurellem.org/vba-clojure/html/total-control.html

  276. http://bakabt.me/159362-umineko-no-naku-koro-ni-music-collection-flac.html

  277. https://bam-dataset.org/

  278. https://hyperdimensionfumo.bandcamp.com/track/how-precious-death-how-senseless-life

  279. https://www.bbc.com/news/uk-scotland-tayside-central-36530120

  280. http://blog.beeminder.com/hieroglyphs/

  281. ⁠, Dmitrii Gerasimov (2019-11-01):

    Table of Contents

    …These days, if you have decent connection, you are seconds away from finding almost any public knowledge in the internet. However, there is another aspect of information: personal and specific to your needs, work and hobbies. It’s your todo list, your private notes, books you are reading. Of course, it’s not that well integrated with the outside world, hence the tooling and experience of interacting with it is very different.

    Some examples:

    • To find something from my Messenger history with a friend, I need to be online, open Facebook, navigate to search and use the interface Facebook’s employees thought convenient (spoiler: it sucks). It’s my information, something that came out from my brain. Why can’t I have it available anywhere, anytime, presented the way I prefer?

    • To find something in my Kobo ebook, I need to reach my device physically and type the query using the virtual keyboard (yep, e-ink lag!). Not a very pleasant experience. It’s something I own and have read. Why does it have to be so hard?

    Such things are pretty frustrating to me, so I’ve been working on making them easier. Search has to be ⁠, fast and as convenient to use as possible. I’ll be sharing some of workflows, tricks and thoughts in this post.

    The post is geared towards using Emacs and Org-mode, but hopefully you’ll find some useful tricks for your current tools and workflow even if you don’t. There is (almost) nothing inherently special about Emacs, I’m sure you can achieve similar workflows in other modern text editors given they are flexible enough.

    Note: throughout the post I will link to my emacs config snippets. To prevent code references from going stale, I use permalinks, but check master branch as well in case of patches or more comments in code.

    What do I search?

    I’ll write about searching in

    • my personal notes, tasks and knowledge repository (this blog included)
    • all digital trace I’m leaving (tweets, internet comments, annotations)
    • chat logs with people
    • books and papers I’m reading
    • code that I’m working on
    • information on the Internet (duh!)
  282. http://bellard.org/jslinux/tech.html

  283. https://bellroy.com/

  284. http://www.berkshirehathaway.com/letters/2011ltr.pdf

  285. https://bibliophilly.library.upenn.edu/viewer.php?id=Ms.%20Codex%201248#page/244/mode/2up

  286. https://bigquery.cloud.google.com/table/fh-bigquery:reddit_comments.2015_05

  287. https://genomebiology.biomedcentral.com/articles/10.1186/gb-2002-3-7-comment2007

  288. ⁠, Xuan Wang, Bo-Wen Zhou, Ting-Ting Yin, Fang-Liang Chen, Ali Esmailizadeh, Melissa M. Turner, Andrei D. Poyarkov, Peter Savolainen, Guo-Dong Wang, Qiaomei Fu, Ya-Ping Zhang (2018-06-19):

    Canine transmissible venereal tumor (CTVT), the oldest known somatic cell line, is a living fossil, originating from cancer cells transmitted from a host to other canids during the mating process7. Clonal origin analyses hints that the original dog infected with CTVT (CTVT founder) came from an ancient sled dog or wolf population. However, the genetic composition of the CTVT founder is still not clear.

    In order to explore this issue, we applied whole genome sequencing (WGS) to two CTVT samples, their corresponding hosts, and 24 additional canids (Supplementary Note). Combined with published WGS data of two CTVT samples and high quality canine WGS data, we constructed a data set containing WGS data of four CTVT samples a 169-individual reference panel composed of worldwide gray wolves (Canis lupus), dogs (Canis lupus familiaris), coyotes (Canis latrans) and golden jackals (Canis aureus) (Supplementary Note, Table S1).

    …Our results reveal that the CTVT founder was more closely related to present-day arctic sled dogs than to any other populations (Figure S6–8), in accordance with very recent results. However, ADMIXTURE analysis showed that the CTVT founder also possessed an ancestral component found predominantly in non-dog populations, a result that we do not observe for any arctic sled dog (Supplementary Note, Figure 1A). Moreover, the CTVT founder did not cluster tightly with arctic sled dogs in the PCA analysis (Figure S7). These results imply that the CTVT founder belonged to a previously unknown arctic dog population that is not represented in the reference panel…In conclusion, our detailed analyses reveal that the CTVT founder came from an arctic sled dog population that possessed introgression from a population related to coyotes, a result that was not known in previous studies. Considering the habitat of coyotes in North America, we propose two hypotheses: (1) The CTVT founder lived in the arctic region of North America. (2) The CTVT founder lived in the arctic region of the Far East, where arctic dogs possessing the introgressed segments migrated through the Bering Strait in an unknown period. Hence, an ancient story of canine admixture is hidden in the genome of a living fossil, the CTVT. To further test our hypotheses of ancient admixture and to better understand the detailed evolutionary history of dogs from the arctic region and Americas, it is crucial to acquire ancient samples in these regions in future work.

  289. https://en.bitcoin.it/wiki/Bets_of_Bitcoin

  290. https://www.blockchain.com/btc/address/1EiVHZnDVjFH6Tic1YmWUSfYmVUnUZdnMU

  291. https://blog.google/products/translate/found-translation-more-accurate-fluent-sentences-google-translate/

  292. https://blog.otoro.net/2016/04/01/generating-large-images-from-latent-vectors/

  293. https://www.bloomberg.com/news/2011-07-28/einstein-on-wall-street-time-money-continuum-commentary-by-mark-buchanan.html

  294. https://www.bls.gov/data/inflation_calculator.htm

  295. ⁠, Sreeram V. Ramagopalan, Marian Knight, George C. Ebers, Julian C. Knight (2007-12-20):

    Objective: To assess the evidence for a genetic basis to magic.

    Design: Literature review.

    Setting: Harry Potter novels of ⁠.

    Participants: Muggles, witches, wizards, and squibs.

    Interventions: Limited.

    Main outcome measures: Family and twin studies, magical ability, and specific magical skills.

    Results: Magic shows strong evidence of heritability, with familial aggregation and concordance in twins. Evidence suggests magical ability to be a quantitative trait. Specific magical skills, notably being able to speak to snakes, predict the future, and change hair colour, all seem heritable.

    Conclusions: A multi-locus model with a dominant gene for magic might exist, controlled epistatically by one or more loci, possibly recessive in nature. Magical enhancers regulating gene expression may be involved, combined with mutations at specific genes implicated in speech and hair colour such as FOXP2 and MCR1.

  296. https://boingboing.net/2010/07/22/ted-chiang-interview.html

  297. http://books.google.com/books?id=7y1jBrpx4loC

  298. http://www.buzzfeed.com/mattlynley/google-reader-died-because-no-one-would-run-it

  299. http://www.buzzricksons.jp/

  300. http://boinngerionn.blog.fc2.com/blog-entry-203.html

  301. http://www.cambridgebrainsciences.com/assets/default/files/pdf/Mehta2000.pdf

  302. https://camelcamelcamel.com/

  303. ⁠, Morgan McGuire (2015):

    [Markdeep is a single-file JavaScript HTML compiler: it can be inserted into a Markdown file, which will automatically render it inside a visiting web browser. It is highly opinionated and featureful, including a wide variety of automatic symbol replacements, ‘admonitions’, embedded ASCII diagrams, calendars, todo task lists, multi-columns, etc.]

    Markdeep is a technology for writing plain text documents that will look good in any web browser, whether local or remote. It supports diagrams, calendars, equations, and other features as extensions of Markdown syntax. Markdeep is free and easy to use. It doesn’t require a plugin or Internet connection. Your document never leaves your machine and there’s nothing to install. Just start writing in your favorite text editor. You don’t have to export, compile, or otherwise process your document. Here’s an example of a text editor and a browser viewing the same file simultaneously:…Markdeep is ideal for design documents, specifications, README files, code documentation, lab reports, blogs, and technical web pages. Because the source is plain text, Markdeep works well with software development toolchains.

    Markdeep was created by Morgan McGuire (Casual Effects) with inspiration from John Gruber’s Markdown and Donald Knuth’s and Leslie Lamport’s LaTeX. Unique features:

    Diagrams · Insert documents into one another · LaTeX equation typesetting and numbering · Table of contents · Reference images and embedded images · Document title and subtitle formatting · Schedules and calendars · Section numbering and references · Figure, listing, and table numbering and references · Smart quotes · Embedded video · CSS stylesheets · Page breaks · En dash, em dash, ×, minus, and degrees · Attributes on links · Unindexed sections · Works in any browser by adding one line to the bottom of a text document · Fallback to ASCII in a browser if you have neither the local file nor Internet access · Optionally process server-side with node.js · Optionally batch process to PDF with headless browser flags · HTML export to static content using ?export in the URL or “Rasterizer”

  304. http://www.catb.org/esr/writings/taouu/html/

  305. https://www.cato-unbound.org/issues/is-more-medicine-better/

  306. https://koeln.ccc.de/archiv/cyphernomicon/chapter10/10.20.html

  307. http://publishing.cdlib.org/ucpressebooks/view?docId=ft6489p0n6&brand=ucpress

  308. https://cdm16630.contentdm.oclc.org/digital/collection/p16630coll2/id/534

  309. https://www.cell.com/cms/attachment/2040464074/2053978065/mmc1.pdf

  310. https://www.census.gov/prod/cen2010/briefs/c2010br-03.pdf

  311. https://ciechanow.ski/cameras-and-lenses/

  312. http://clickotron.com/

  313. https://clinicaltrials.gov/ct2/show/record/NCT01257867

  314. http://tryr.codeschool.com/

  315. http://www.cogmed.com/

  316. http://cognitivefun.net/talk/post/22963

  317. http://cogprints.org/371/3/148.pdf

  318. http://www.cogtest.com/coglib_demtest.html

  319. https://colab.research.google.com/drive/1xLAsfefpG1Hom7JZYbvdROHugMBNBrmR

  320. https://www.coursera.org/course/dataanalysis

  321. https://cran.r-project.org/web/packages/orderstats/index.html

  322. https://creativecommons.org/licenses/by-nc/3.0/us/

  323. http://www.cryonicscalculator.com/

  324. http://cryptome.org/jya/digicrash.htm

  325. https://programmablesearchengine.google.com/about/cse?cx=009114923999563836576:1eorkzz2gp4

  326. https://ctan.org/pkg/marginnote

  327. https://www.dafont.com/kanzlei.font

  328. ⁠, Dan Luu (2017-10-16; technology⁠, cs⁠, design):

    [Dan Luu continues his investigation of why computers feel so laggy and have such high latency compared to old computers (⁠, ⁠, ⁠, cf text editor analysis).

    He measures 21 keyboard latencies using a logic analyzer, finding a range of 15–60ms (!), representing a waste of a large fraction of the available ~100–200ms latency budget before a user notices and is irritated (“the median keyboard today adds as much latency as the entire end-to-end pipeline as a fast machine from the 70s.”). The latency estimates are surprising, and do not correlate with advertised traits. They simply have to be measured empirically.]

    We can see that, even with the limited set of keyboards tested, there can be as much as a 45ms difference in latency between keyboards. Moreover, a modern computer with one of the slower keyboards attached can’t possibly be as responsive as a quick machine from the 70s or 80s because the keyboard alone is slower than the entire response pipeline of some older computers. That establishes the fact that modern keyboards contribute to the latency bloat we’ve seen over the past forty years…Most keyboards add enough latency to make the user experience noticeably worse, and keyboards that advertise speed aren’t necessarily faster. The two gaming keyboards we measured weren’t faster than non-gaming keyboards, and the fastest keyboard measured was a minimalist keyboard from Apple that’s marketed more on design than speed.

  329. https://us.dantelabs.com/collections/best-seller/products/whole-genome-sequencing-wgs-full-dna-analysis

  330. ⁠, Dan Wang (2017-06-25; sociology  /​ ​​ ​preference-falsification):

    …competition is fiercer the more that competitors resemble each other. When we’re not so different from people around us, it’s irresistible to become obsessed about beating others.

    It’s hard to construct a more perfect incubator for mimetic contagion than the American college campus. Most 18-year-olds are not super differentiated from each other. By construction, whatever distinctions any does have are usually earned through brutal, zero-sum competitions. These tournament-type distinctions include: SAT scores at or near perfection; being a top player on a sports team; gaining master status from chess matches; playing first instrument in state orchestra; earning high rankings in Math Olympiad; and so on, culminating in gaining admission to a particular college. Once people enter college, they get socialized into group environments that usually continue to operate in zero-sum competitive dynamics. These include orchestras and sport teams; fraternities and sororities; and many types of clubs. The biggest source of mimetic pressures are the classes. Everyone starts out by taking the same intro classes; those seeking distinction throw themselves into the hardest classes, or seek tutelage from star professors, and try to earn the highest grades.

    There’s very little external intermediation, instead all competitive dynamics are internally mediated…Once internal rivalries are sorted out, people coalesce into groups united against something foreign. These tendencies help explain why events on campus so often make the news—it seems like every other week we see some campus activity being labeled a “witch hunt”, “riot”, or something else that involves violence, implied or explicit. I don’t care to link to these events, they’re so easy to find. It’s interesting to see that academics are increasingly becoming the target of student activities. The terror devours its children first, who have tolerated or fanned mimetic contagion for so long.

    …I’ll end with a quote from I See Satan Fall Like Lightning: “Mimetic desire enables us to escape from the animal realm. It is responsible for the best and the worst in us, for what lowers us below the animal level as well as what elevates us above it. Our unending discords are the ransom of our freedom.”

  331. http://darkdata.bc.edu/

  332. https://darknetlive.com/arrested-darknet-vendors/

  333. http://darwin-online.org.uk/content/frameset?itemID=F1548.1&viewtype=text&pageseq=1

  334. http://www.datascienceassn.org/sites/default/files/A%20Few%20Useful%20Things%20to%20Know%20about%20Machine%20Learning.pdf

  335. https://www.davidsongifted.org/Search-Database/entry/A10489

  336. http://popcon.debian.org/

  337. http://www.deeplearningbook.org/contents/rnn.html

  338. https://deepmind.com/blog/specifying-ai-safety-problems/

  339. https://derpibooru.org/1707469

  340. https://discord.gg/7RgdJZy

  341. https://www.discovermagazine.com/2011/mar/24-the-brain-the-trouble-with-teens/article_view?b_start:int=1&-C=

  342. http://www.doc88.com/p-397166703921.html

  343. https://docs.google.com/forms/d/e/1FAIpQLSeT3GIg-pSwzDFAfNaqE-MzfJEtD0HghN_Vma68OZJtz1Pztg/viewform

  344. http://docsdrive.com/pdfs/academicjournals/jpt/2008/425-434.pdf

  345. http://dx.doi.org/10.1037/a0014104

  346. https://danbooru.donmai.us/wiki_pages/4920

  347. https://drive.google.com/drive/folders/1NY3MTkOSodz_5eCkjtoHUGYSulR25QGU

  348. https://www.drmaciver.com/2019/05/how-to-do-hard-things/

  349. https://www.dropbox.com/s/qyvgn4y2iqs6g3n/setsugen-%E5%A5%8F%E4%B9%B1%E7%B5%B5%E5%B7%BB-%E5%A3%B1%E3%80%81%E6%84%9F%E6%83%85%E3%81%AE%E6%91%A9%E5%A4%A9%E6%A5%BCemotionalskyscraper.ogg

  350. https://dl.dropboxusercontent.com/u/280585369/music/%E3%82%82%E3%81%B5%EF%BC%A0tinkle-pop-6%E4%BA%BA%E3%81%AE%E5%90%9B%E3%81%A8%E5%83%95%E3%81%AE%E6%AD%8C-%E6%BA%80%E3%81%9F%E3%81%95%E3%82%8C%E3%81%9F%E3%81%A8%E3%81%8D.ogg

  351. https://dspace.mit.edu/openaccess-disseminate/1721.1/116112

  352. http://www.dtic.mil/dtic/tr/fulltext/u2/a229000.pdf

  353. http://dual-n-back.com/nback.thml

  354. https://duckduckgo.com/bang#bangs-list

  355. https://www.duolingo.com/

  356. http://dwarffortresswiki.org/index.php/User:BaronW#The_Almighty_Dwarven_Calculator

  357. http://www.ecologyandsociety.org/vol9/iss1/art6/main.html

  358. https://www.econlib.org/archives/2016/10/what_do_crimina.html

  359. https://www.economist.com/news/science-and-technology/21567879-yet-another-study-suggests-sperm-numbers-are-falling-rich-countries-countdown

  360. http://www.econtalk.org/archives/2012/04/autor_on_disabi.html

  361. https//www.edge.org/conversation/-the-book-of-revelation-prophecy-and-politicsedge-master-class-2011

  362. http://eepurl.com/cGAN7L

  363. https://elifesciences.org/articles/31157

  364. http://www.emacswiki.org/emacs/BrowseUrl

  365. http://www.epjournal.net/blog/2011/08/glucose-is-not-willpower-fuel/

  366. https://www.equator-network.org/reporting-guidelines/

  367. https://www.equestriadaily.com/search/label/Music

  368. https://equilibriabook.com/molochs-toolbox/

  369. https://ergodox-ez.com/

  370. https://eric.ed.gov/?id=ED176997

  371. http://de1.erowid.org/experiences/subs/exp_Melatonin.html

  372. http://www.esajournals.org/doi/full/10.1890/14-0661.1

  373. https://escholarship.org/content/qt1kn271xs/qt1kn271xs.pdf

  374. https://eurekamag.com/

  375. http://www.europeanneuropsychopharmacology.com/article/S0924-977X%2817%2930019-6/fulltext

  376. https://forum.evageeks.org/post/637739/Books-Shizo-Parano/#637739

  377. http://www.evamonkey.com/after-the-end/

  378. https://everything2.com/title/You+have+a+sad+feeling+for+a+moment%252C+then+it+passes

  379. http://examine.com/supplements/DMAE/

  380. https://explorabl.es/

  381. http://f1000research.com/articles/3-82/v1

  382. https://ai.facebook.com/blog/mapping-roads-through-deep-learning-and-weakly-supervised-training

  383. https://mlp.fandom.com/wiki/Do_Princesses_Dream_of_Magic_Sheep%3F

  384. https://www.fanfiction.net/s/5782108/65/Harry_Potter_and_the_Methods_of_Rationality

  385. http://www.farnamstreetblog.com/2014/02/decision-journal/

  386. http://akenney.fastmail.fm/works/9.61Paper.pdf

  387. http://feeds.feedburner.com/longbets

  388. ⁠, 15 & the Pony Preservation Project (2020-03-06):

    [NN TTS service demonstrating results from custom DL research project by 15 for generating natural high-quality voices of characters with minimal data/​​​​few-shot learning; available voices include GLaDOS from Portal and especially high-quality My Little Pony: Friendship Is Magic voices (currently: Fluttershy & Twilight Sparkle); demos: 1⁠/​​​​2⁠.

    The MLP:FiM voices are trained on a large dataset constructed by the 4chan crowdsourced project “Pony Preservation Project”, begun ~2019. PPP has crowdsourced parsed audio and hand-written transcriptions of all dialogue for all character from all 9 MLP:FiM seasons, the movie, the spinoffs, and various other things voiced by the same voice actresses in case that might help, while processing to remove noise or using ‘leaked’ original data from Hasbro for higher quality still.]

    This is a text-to-speech tool that you can use to generate 44.1 kHz voices of various characters. The voices are generated in real time using multiple audio synthesis algorithms and customized deep neural networks trained on very little available data (between 30 and 120 minutes of clean dialogue for each character). This project demonstrates a substantial reduction in the amount of audio required to realistically clone voices while retaining their affective prosodies.

    I plan to keep this tool up gratis and ad-free indefinitely. This website is intended for strictly non-commercial use.

    Thanks to the MIT Computer Science & Artificial Intelligence Laboratory (CSAIL) for providing the initial funding that kickstarted this project two years ago. Further thanks to the Julia Lab, Lincoln Lab, and the Media Lab.

    Special shoutouts go to 4chan’s /mlp/ and the anons who have collectively spent hundreds of hours collecting, cleaning, and organizing clips of dialogue taken from the show My Little Pony: Friendship Is Magic. Honorable mention to /g/ for some entertaining speculations.

    And of course, nothing but the utmost respect to the voice actors who originally voiced the characters.

  389. http://www.fightaging.org/archives/2011/05/more-on-body-temperature-and-calorie-restriction.php

  390. https://www.filfre.net/2012/07/the-dennis-wheatley-crime-dossiers/

  391. https://www.find-more-books.com/

  392. http://flashcarddb.com/cardsets

  393. http://www.flashcardexchange.com

  394. https://flatisjustice.moe/TADNE

  395. https://secure.flickr.com/photos/mharrsch/249083991/

  396. https://yp.flutterguy.org/

  397. http://folding.stanford.edu/English/FAQ-Diseases

  398. https://folkrnn.org/

  399. https://www.frontiersin.org/Journal/10.3389/fnsys.2014.00152/full

  400. https://fullfrontal.moe/khara-10-part-2/

  401. ⁠, Arfafax (2020-06-01):

    [Google Colab notebook for interactive editing faces generated by the face model, using Ganspace to reverse-engineer the latent encoding and allow control of specific visual attributes of faces.]

  402. https://gallica.bnf.fr/ark:/12148/bpt6k851127r/

  403. https://www.artbreeder.com/browse

  404. http://libgen.rs/scimag/

  405. http://www.genetics.org/content/202/3/869

  406. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4015143/

  407. http://www.getlamp.com/

  408. https://make.girls.moe

  409. ⁠, Vivek Singh (2019-02-22):

    A few weeks ago, we announced a radical experiment in Open Source Funding. Using the matching method outlined in —a paper by Glen Weyl, Vitalik Buterin, and Zoë Hitzig—we announced a $25K fund to match any contributions made to 25 Ethereum infrastructure projects.

    Gitcoin’s CLR Matching, By The Numbers

    • The top three projects in matching funding for this round were Prysmatic Labs, Moloch DAO, and Uniswap
    • $13,242 was contributed by 132 unique contributors across 26 projects
    • The top 10 projects all received over $1,000 in matching donations from the CLR fund
    • A total of $38,242 was contributed to Ethereum OSS infrastructure in two weeks

    …We are encouraged by the results made in the first round of CLR matching and are hopeful for the emergence of new mechanisms to enable funding to public goods. We’ll explore a few of these in future rounds, and are especially interested in inflation funding mechanisms to fund public infrastructure.

  410. https://github.com/fpco/github

  411. ⁠, C. Daniel Freeman, Luke Metz, David Ha (2019-10-29):

    Much of model-based reinforcement learning involves learning a model of an agent’s world, and training an agent to leverage this model to perform a task more efficiently. While these models are demonstrably useful for agents, every naturally occurring model of the world of which we are aware—eg., a brain—arose as the byproduct of competing evolutionary pressures for survival, not minimization of a supervised forward-predictive loss via gradient descent. That useful models can arise out of the messy and slow optimization process of evolution suggests that forward-predictive modeling can arise as a side-effect of optimization under the right circumstances. Crucially, this optimization process need not explicitly be a forward-predictive loss. In this work, we introduce a modification to traditional reinforcement learning which we call observational dropout, whereby we limit the agents ability to observe the real environment at each timestep. In doing so, we can coerce an agent into learning a world model to fill in the observation gaps during reinforcement learning. We show that the emerged world model, while not explicitly trained to predict the future, can help the agent learn key skills required to perform well in its environment.

    [Image caption: “Our agents are only given infrequent observations of the real environment. As a side effect for optimizing performance in this setting, a”world model" emerges. We show the true dynamics in color, with full saturation denoting frames the policy can see. The black and white outline shows the state of the emergent world model. These world model exhibits similar, but not identical dynamics to forward predictive models but only model “important” aspects of the environment."]

  412. https://gitlab.com/cryptsetup/cryptsetup/issues/77

  413. https://blog.givewell.org/2016/01/19/the-importance-of-gold-standard-studies-for-consumers-of-research/

  414. ⁠, Whitney Kimball (Gizmodo) (2020-05-17):

    Much of the fun of internet drama comes from its frivolousness, but sometimes an online shitfest points to something bigger. Last week, the AI-powered furry art site did just that, igniting a fandom firestorm while also highlighting an important debate about digital art. Trained on more than 55,000 images pulled (without permission) from a furry art forum, the algorithm was a simple case of art theft to some. For others, it was a chance to break out the popcorn. But legal scholars who spoke with Gizmodo said the conflict raises thorny questions about ownership in the age of AI—questions that may ultimately have to be answered in court.

    …At least one person tried (and failed) to find proof that the algorithm was copying images from e621.net outright. And within days, the entire site was slapped with a DMCA copyright infringement complaint. (The company whose name the DMCA was issued in, according to Arfa, denied filing the notice and requested it be withdrawn.) Some degree of backlash is understandable. Furry fandom has long been a close-knit community of independent creators supported by individual commissions. A project aimed at mass-producing fursonas—using original art as training material, no less—could be seen as a threat to creators’ livelihood. Some commenters accused Arfa of disrespect and asked for the choice to opt out of the project. Others complained that their work had been uploaded to e621 without their permission in the first place.

  415. http://permalink.gmane.org/gmane.science.linguistics.wikipedia.english/110790

  416. https://www.gnxp.com/WordPress/2018/04/20/so-merfolk-are-a-real-thing-now/

  417. https://goo.gl/1AfwZG

  418. https://www.goodreads.com/interviews/show/1169.Colson_Whitehead

  419. http://www.google-melange.com/gsoc/homepage/google/gsoc2011

  420. https://programmablesearchengine.google.com/about/cse/publicurl?cx=009114923999563836576:dv0a4ndtmly

  421. http://www.google.com/search?q=gwern+site%3Ahttp%3A%2F%2Fwww.haskell.org%2Fpipermail%2Fxmonad%2F

  422. https://patentimages.storage.googleapis.com/57/53/22/91b8a6792dbb1e/US20180204116A1.pdf

  423. http://google-summer-of-code-2009-haskell.googlecode.com/files/IsaacA_Dupree.tar.gz

  424. http://www.goproblems.com

  425. http://gptprompts.wikidot.com/logic:math

  426. http://philip.greenspun.com/teaching/teaching-software-engineering

  427. http://stats.grok.se/en/201109/Accountancy

  428. http://groups.google.com/group/isrs-support

  429. http://www.gutenberg.org/files/9105/9105-h/9105-h.htm

  430. https://guzey.com

  431. 20130103-20130702-gwern.net-analytics.pdf: “Gwern.net: Google Analytics Semi-Annual Traffic Report (20130103-20130702)”⁠, Gwern Branwen

  432. https://halshs.archives-ouvertes.fr/halshs-01948311/document

  433. https://www.harney.com/products/bancha

  434. https://wiki.haskell.org/Haskell

  435. https://babel.hathitrust.org/cgi/pt?id=mdp.39015039380632&view=1up&seq=333

  436. https://haveibeenpwned.com/

  437. http://hpmor.com/

  438. https://www.html-tidy.org/

  439. https://almanac.httparchive.org/en/2019/

  440. ⁠, Teven Le Scao (Hugging Face) (2020-06-08):

    [Discussion of DL scaling laws and how big = better, with interactive graphs to help visualize the multi-way relationship between dataset / model / validation-loss / FLOPS.]

    Research at Hugging Face also leverages this phenomenon, and we’ve combined it with GPU speed estimations to ensure model size is just right for the compute budget of the experiment (when in doubt, it’s bigger than you think!). This blog post will show how this impacts architecture decisions on a standard language modeling benchmark: we replicate the 14-layer state-of-the-art result from Zhang et al.’s paper without any hyper-parameter optimization and saving 25% of training time. We also estimate that the 18-layer model from the same paper trained for an order of magnitude too many training steps. Wanna play with our demo before reading? Just click here!

    1. There is an optimal time to stop training (and it’s earlier than you think)

    2. GPUs are optimized for large, wide models

    3. Demonstration on a language modeling task: Wikitext-103

    4. Takeaways

      • Big models are surprisingly efficient!
      • Training until convergence is not efficient at all.
      • Benchmarking smaller-scale runs allows us to predict model performance and time for production-scale models.
      • Using larger models stopped earlier and optimizing model size for speed lowers training costs.
  441. https://www.humanprogress.org/article.php?p=631

  442. http://prize.hutter1.net/

  443. https://eprint.iacr.org/2018/601.pdf

  444. ⁠, Maciej Ceglowski (2010-06-03):

    [Scott’s Antarctic expedition in 1911 was plagued by the disease scurvy, despite its having been “conquered in 1747, when the Scottish physician James Lind proved in one of the first controlled medical experiments that citrus fruits were an effective cure for the disease.” How it all went wrong would make a case study for a philosophy of science class.

    The British Admiralty switched their scurvy cure from lemon juice to lime juice in 1860. The new cure was much less effective, but by that time advances in technology meant that most sea voyages were so short that there was little or no danger of scurvy anyway. So poor Scott’s expedition, as well as applying ‘state-of-the-art’ (ie. wrong) cures, were falling back on a ‘tried-and-true’ remedy that in fact had been largely ineffective already for 50 years… without anyone noticing.]

    An unfortunate series of accidents conspired with advances in technology to discredit the cure for scurvy. What had been a simple dietary deficiency became a subtle and unpredictable disease that could strike without warning. Over the course of fifty years, scurvy would return to torment not just Polar explorers, but thousands of infants born into wealthy European and American homes. And it would only be through blind luck that the actual cause of scurvy would be rediscovered, and vitamin C finally isolated, in 1932.

    …So when the Admiralty began to replace lemon juice with an ineffective substitute in 1860, it took a long time for anyone to notice. In that year, naval authorities switched procurement from Mediterranean lemons to West Indian limes. The motives for this were mainly colonial—it was better to buy from British plantations than to continue importing lemons from Europe. Confusion in naming didn’t help matters. Both “lemon” and “lime” were in use as a collective term for citrus, and though European lemons and sour limes are quite different fruits, their Latin names (citrus medica, var. limonica and citrus medica, var. acida) suggested that they were as closely related as green and red apples. Moreover, as there was a widespread belief that the antiscorbutic properties of lemons were due to their acidity, it made sense that the more acidic Caribbean limes would be even better at fighting the disease.

    In this, the Navy was deceived. Tests on animals would later show that fresh lime juice has a quarter of the scurvy-fighting power of fresh lemon juice. And the lime juice being served to sailors was not fresh, but had spent long periods of time in settling tanks open to the air, and had been pumped through copper tubing. A 1918 animal experiment using representative samples of lime juice from the navy and merchant marine showed that the ‘preventative’ often lacked any antiscorbutic power at all.

    By the 1870s, therefore, most British ships were sailing without protection against scurvy. Only speed and improved nutrition on land were preventing sailors from getting sick.

    …In the course of writing this essay, I was tempted many times to pick a villain. Maybe the perfectly named Almroth Wright, who threw his considerable medical reputation behind the ptomaine theory and so delayed the proper re-understanding of scurvy for many years. Or the nameless Admiralty flunky who helped his career by championing the switch to West Indian limes. Or even poor Scott himself, sermonizing about the virtues of scientific progress while never conducting a proper experiment, taking dreadful risks, and showing a most unscientific reliance on pure grit to get his men out of any difficulty.

    But the villain here is just good old human ignorance, that master of disguise. We tend to think that knowledge, once acquired, is something permanent. Instead, even holding on to it requires constant, careful effort.

  445. http://www.ieee-security.org/TC/SP2013/papers/4977a080.pdf

  446. http://ieet.org/index.php/IEET/more/hughes20121016

  447. http://www.ietf.org/rfc/rfc5005.txt

  448. http://www.imagemagick.org/Usage/resize/

  449. https://www.imdb.com/name/nm0663048/

  450. https://i.imgur.com/jzZKreU.png

  451. http://www.incompleteideas.net/book/the-book-2nd.html

  452. http://home.inklingmarkets.com/expiring/markets

  453. https://intelligence.org/files/AIPosNegFactor.pdf

  454. http://www.intrade.com/jsp/intrade/help/index.jsp?page=general.html%23fees

  455. https://inventingabstraction.tumblr.com/post/42450192391/we-heard-that-the-original-image-of-the-1930

  456. http://iqcomparisonsite.com/

  457. http://www.iqout.com/

  458. http://www.iqtest.com/

  459. http://scielo.isciii.es/scielo.php?script=sci_arttext&pid=S0212-97282016000300004

  460. http://www.isfdb.org/cgi-bin/pl.cgi?261005

  461. http://www.jacurutu.com/viewtopic.php?f=21&t=1112#p34878

  462. https://jamanetwork.com/journals/jamaoncology/fullarticle/2499777

  463. https://jasoncrawford.org/

  464. http://www.jneurosci.org/content/32/12/4156.full

  465. https://www.joelonsoftware.com/2000/06/03/strategy-letter-iii-let-me-go-back/

  466. http://www.johndcook.com/blog/2010/12/09/does-typing-speed-matter/

  467. http://johnmacfarlane.net/pandoc/

  468. http://www.journalofvision.org/content/12/9/177.abstract?sid=8ab3ce55-344f-4b31-a6c1-c711d060237a

  469. http://www.jstatsoft.org/v35/i03/paper

  470. https://www.jstor.org/stable/pdf/3803390.pdf

  471. http://www.jwz.org/gruntle/design.html

  472. https://www.kaggle.com/alamson/safebooru

  473. https://training.kalzumeus.com/newsletters/archive/do-not-end-the-week-with-nothing

  474. http://dev.kanotype.net:8003/deepdanbooru/

  475. https://keras.io/

  476. https://www.kickstarter.com/projects/pijuice/pijuice-a-portable-project-platform-for-every-rasp

  477. https://www.kill-the-newsletter.com/

  478. https://kk.org/thetechnium/better-than-fre/

  479. https://knightcolumbia.org/page/about-the-knight-institute

  480. https://landing.google.com/sre/books/

  481. https://leaderboard.allenai.org/break_high_level/submissions/public

  482. http://learnyouahaskell.com/functors-applicative-functors-and-monoids#monoids

  483. https://www.ledger-cli.org

  484. https://leme.me/verah/mp3/?C93/Lost%20Garden%20%E2%80%94%20ENIGMATIC%20LINER%20%5BMP3-V0%5D%5BC93%5D#trk6

  485. https://library.bz/main/upload/

  486. http://libgen.org/search.php?req=%22wheel+of+time%22

  487. https://www.librarything.com/work/13068

  488. ⁠, Andrew E. Snyder-Beattie, ⁠, K. Eric Drexler, Michael B. Bonsall (2020-11-19):

    It is unknown how abundant extraterrestrial life is, or whether such life might be complex or intelligent. On Earth, the emergence of complex intelligent life required a preceding series of evolutionary transitions such as abiogenesis, eukaryogenesis, and the evolution of sexual reproduction, multicellularity, and intelligence itself. Some of these transitions could have been extraordinarily improbable, even in conducive environments. The emergence of intelligent life late in Earth’s lifetime is thought to be evidence for a handful of rare evolutionary transitions, but the timing of other evolutionary transitions in the fossil record is yet to be analyzed in a similar framework.

    Using a simplified that combines uninformative priors and the timing of evolutionary transitions, we demonstrate that expected evolutionary transition times likely exceed the lifetime of Earth, perhaps by many orders of magnitude. Our results corroborate the original argument suggested by Brandon Carter that intelligent life in the Universe is exceptionally rare, assuming that intelligent life elsewhere requires analogous evolutionary transitions. Arriving at the opposite conclusion would require exceptionally conservative priors, evidence for much earlier transitions, multiple instances of transitions, or an alternative model that can explain why evolutionary transitions took hundreds of millions of years without appealing to rare chance events.

    Although the model is simple, it provides an initial basis for evaluating how varying biological assumptions and fossil record data impact the probability of evolving intelligent life, and also provides a number of testable predictions, such as that some biological paradoxes will remain unresolved and that planets orbiting M dwarf stars are uninhabitable.

    [Previous work:

  489. https://listudy.org

  490. http://longbets.org/285/

  491. http://reprints.longform.org/perry-once-upon-a-jihad

  492. https://longreads.com/2018/09/27/queens-of-infamy-the-early-trials-of-catherine-de-medici/

  493. https://lwn.net/Articles/766374/

  494. https://magicemail.io/

  495. https://www.mail-archive.com/cryptography@metzdowd.com/msg10015.html

  496. https://marginalrevolution.com/marginalrevolution/2013/06/the-oocyte-cartel.html

  497. http://mathoverflow.net/questions/53122/mathematical-urban-legends/53168#comment132000_53168

  498. https://www.mayoclinicproceedings.org/article/S0025-6196%2813%2900405-9/fulltext

  499. http://mc-stan.org/

  500. https://www.mdpi.com/2076-3425/8/3/46/htm

  501. http://www.mediawiki.org/wiki/Extension:ConfirmEdit

  502. ⁠, Chloe Fawns-Ritchie, Ian J. Deary (2019-07-15):

    UK is a health resource with data from over 500,000 adults. The participants have been assessed on cognitive function since baseline. The cognitive tests in UK Biobank are brief and bespoke, and are administered without supervision on a touchscreen computer. Psychometric information on the tests is limited. The present study examined their concurrent validity and short-term test-retest reliability. A sample of 160 participants (mean age = 62.59, SD = 10.24) completed the UK Biobank cognitive assessment and a range of well-validated cognitive tests (‘reference tests’). Fifty-two participants returned 4 weeks later to repeat the UK Biobank tests. Correlations were calculated between UK Biobank tests and the reference tests. Four-week test-retest correlations were calculated for UK Biobank tests. UK Biobank cognitive tests showed a range of correlations with their respective reference tests, i.e. those tests that are thought to assess the same underlying cognitive ability (mean Pearson r = 0.53, range = 0.22 to 0.83, p≤.005). Four-week test-retest reliability of the UK Biobank tests were moderate-to-high (mean Pearson r = 0.55, range = 0.40 to 0.89, p≤.003). Despite the brief, non-standard nature of the UK Biobank cognitive tests, some showed substantial concurrent validity and test-retest reliability. These psychometric results provide currently-lacking information on the validity of the UK Biobank cognitive tests.

  503. http://www.meltingasphalt.com/hallucinated-gods/

  504. https://www.metaculus.com/questions/3479/when-will-the-first-artificial-general-intelligence-system-be-devised-tested-and-publicly-known-of/

  505. https://www.metafilter.com/91797/working-working-memory-with-dual-nback#3081757

  506. http://www.metafor-project.org/

  507. https://www.millionshort.com/

  508. http://mnemosyne-proj.org/card-sets

  509. https://www.motherjones.com/politics/2016/06/cca-private-prisons-corrections-corporation-inmates-investigation-bauer/

  510. https://developer.mozilla.org/en-US/docs/Web/CSS/@media/prefers-color-scheme

  511. https://mruniversity.com/courses

  512. https://myanimelist.net/animelist/gwern

  513. ⁠, Nature (2020-07-08):

    Could spreading basalt dust on farmers’ fields help to remove atmospheric carbon? A large multidisciplinary team of scientists is confident it could, and that doing so could boost crop yields and soil health at the same time…The team is also carrying out field trials in four countries—the only such trials yet. The authors have told Nature that preliminary results suggest the theory is holding up. The application of 20 tonnes of basalt dust to a half-hectare UK plot boosted CO2 removal by 40% compared with that seen on an untreated plot, and by 15% in another trial, which spread dust over oil-palm plantations in Malaysia. The early results also indicate that adding basalt boosted yields in these and other crops.

    …But, like many promising technological fixes, spreading basalt dust across the world’s agricultural fields could prove more complicated than it first seems. Researchers must answer a host of pressing questions about the economic costs and environmental impacts. And there are potential questions for regulators, too.

    …With the dangers of climate change becoming more apparent each year, countries must continue to pursue the aggressive action that will be required to meet the goals of the 2015 Paris climate agreement. Carbon-removal technologies cannot be a substitute for such action. But it is becoming clear that if humanity is to limit global warming to 1.5–2 °C above pre-industrial levels, it must pursue every promising idea.

  514. http://www.nber.org/papers/w20366.pdf

  515. https://ncase.me/remember/

  516. http://publicsearch.ndcourts.gov/default.aspx

  517. https://www.nearlyfreespeech.net

  518. https://www.nejm.org/doi/full/10.1056/NEJM199605023341801

  519. http://neuralnetworksanddeeplearning.com/chap6.html

  520. http://www.newegg.com/Product/Product.aspx?Item=N82E16834314853

  521. https://www.newyorker.com/science/maria-konnikova/practice-doesnt-make-perfect

  522. https://www.nickbostrom.com/ethics/statusquo.pdf

  523. http://www.nicovideo.jp/watch/sm23047550

  524. ⁠, Michael A. ⁠, Jasmine Ide, Sarah E. Masters, Thomas J. Carew (2002-01):

    In Aplysia, three distinct phases of memory for sensitization can be dissociated based on their temporal and molecular features. A single training trial induces short-term memory (STM, lasting <30 min), whereas five trials delivered at 15-min intervals induces both intermediate-term memory (ITM, lasting >90 min) and long-term memory (LTM, lasting >24 h). Here, we explore the interaction of amount and pattern of training in establishing ITM and LTM by examining memory for sensitization after different numbers of trials (each trial = one tail shock) and different patterns of training (massed vs. spaced). Under spaced training patterns, two trials produced STM exclusively, whereas four or five trials each produced both ITM and LTM. Three spaced trials failed to induce LTM but did produce an early decaying form of ITM (E-ITM) that was statistically-significantly shorter and weaker in magnitude than the late-decaying ITM (L-ITM) observed after four to five trials. In addition, E-ITM was induced after three trials with both massed and spaced patterns of training. However, L-ITM and LTM after four to five trials require spaced training: Four or five massed trials failed to induce LTM and produced only E-ITM. Collectively, our results indicate that in addition to three identified phases of memory for sensitization—STM, ITM, and LTM—a unique temporal profile of memory, E-ITM, is revealed by varying either the amount or pattern of training.

  525. ⁠, José Luis (2020-06-16):

    It has recently been found possible to estimate age, mortality risk, or general health by looking merely at the epigenome. The models used to do so are referred to as ⁠.

    Epigenetic clocks are increasingly becoming a popular choice for scientists in the field of aging research to measure the putative efficacy of anti-aging interventions. They may make it possible to get results before full curves are available, and they could serve, at least seemingly, as a replacement for a host of other biomarkers. I recommend reading the introductory sections of The Longevity FAQ as well as those about epigenetics before reading this post as it gives some more context.


    Even with a small number of the CpGs of the epigenome measured, it has been possible to construct clocks that accurately track age and health. We still don’t know exactly why the clocks work, just that they do. There is some interesting evidence pointing out to at least part of the pattern seem in the aged epigenome being causal, not just a reflection of the overall condition of the tissue or organism, so we may soon see the epigenome becoming a target for novel drugs. If you want to continue reading about this, Bell et. al 2019’s review (from where I extract the table below) and are the best starting points.

  526. http://papers.nips.cc/paper/5421-deep-learning-for-real-time-atari-game-play-using-offline-monte-carlo-tree-search-planning.pdf

  527. http://www.nixnote.org/

  528. http://arbtt.nomeata.de/

  529. http://www.nongnu.org/rdiff-backup/

  530. http://norvig.com/prayer.html

  531. ⁠, Dan Charles (NPR) (2019-10-30):

    Foods go in and out of style. Few of them, though, have gone through as dramatic a renaissance in their reputation as Brussels sprouts. For many years, they were scorned. Even Steve Bontadelli admits it, and he makes his living growing them. “A lot of people of my generation hated them”, he says. “Their moms boiled them and made them even stinkier.” Bontadelli’s farm is near Santa Cruz, Calif., where the weather is perfect for growing this vegetable. “We actually had a Brussels sprouts festival here for about 10 years”, he says. “And we got a lot of free press out of the deal, because people couldn’t believe that you’d have a festival for Brussels sprouts.” What’s worse, they even deserved their bad reputation. “They were just very bitter; a very strong bitter taste”, Bontadelli says.

    This all started to change in the 1990s, and it began in the Netherlands, where Brussels sprouts have a simpler name: spruitjes. A Dutch scientist named Hans van Doorn, who worked at the seed and chemical company Novartis (the seed part is now called Syngenta), figured out exactly which chemical compounds in spruitjes made them bitter. At that point, the small handful of companies that sell Brussels sprouts seeds started searching their archives, looking for old varieties that happen to have low levels of the bitter chemicals.

    …There are hundreds of these old varieties. The companies grew them in test plots, and they did, in fact, find some that weren’t as bitter. They cross-pollinated these old varieties with modern, high-yielding ones, trying to combine the best traits of old and new spruitjes. It took many years. But it worked. “From then on, the taste was much better. It really improved”, Sintenie says.

    Then word spread in the professional culinary scene. It took off mainly in the United States, not in Europe. Shannan Troncoso remembers hearing, about a decade ago, that celebrity chef David Chang was doing amazing things with Brussels sprouts and bacon at his restaurant Momofuku, in New York. Then she encountered some crispy fried Brussels sprouts at a restaurant in San Francisco. “It was so good, I was like, I can figure this out! And I can introduce this back into my area”, she says…Demand is booming; farmers are getting four or five times more money than they did a decade ago for their crop. “My dad, his jaw would just drop”, Bontadelli says. “He’d ask me every day, ‘What’s the price, what’s the price?’ Because he’d been in the business his whole life. His eyes would just pop out when I’d tell him. He couldn’t believe it.” Bontadelli says that there were only about 2,500 acres in the whole country planted with Brussels sprouts just a few years ago. Today, there are 10,000 acres of Brussels sprouts in the U.S., and fields are getting planted in Mexico, too—just so people can get their Brussels sprouts year-round.

  532. ⁠, Andy Matuschak, Michael Nielsen (2019-10):

    [Long writeup by Andy Matuschak and Michael Nielsen on experiment in integrating spaced repetition systems with a tutorial on quantum computing, Quantum Country: Quantum Computing For The Very Curious By combining explanation with spaced testing, a notoriously thorny subject may be learned more easily and then actually remembered—such a system demonstrating a possible ‘tool for thought’. Early results indicate users do indeed remember the quiz answers, and feedback has been positive.]

    Part I: Memory systems

    • Introducing the mnemonic medium
    • The early impact of the prototype mnemonic medium
    • Expanding the scope of memory systems: what types of understanding can they be used for?
    • Improving the mnemonic medium: making better cards
    • Two cheers for mnemonic techniques
    • How important is memory, anyway?
    • How to invent Hindu-Arabic numerals?

    Part II: Exploring tools for thought more broadly:

    • Mnemonic video

    • Why isn’t there more work on tools for thought today?

    • Questioning our basic premises

      • What if the best tools for thought have already been discovered?
      • Isn’t this what the tech industry does? Isn’t there a lot of ongoing progress on tools for thought?
      • Why not work on AGI or BCI instead?
    • Executable books

      • Serious work and the aspiration to canonical content
      • Stronger emotional connection through an inverted writing structure

    Summary and Conclusion

    … in Quantum Country an expert writes the cards, an expert who is skilled not only in the subject matter of the essay, but also in strategies which can be used to encode abstract, conceptual knowledge. And so Quantum Country provides a much more scalable approach to using memory systems to do abstract, conceptual learning. In some sense, Quantum Country aims to expand the range of subjects users can comprehend at all. In that, it has very different aspirations to all prior memory systems.

    More generally, we believe memory systems are a far richer space than has previously been realized. Existing memory systems barely scratch the surface of what is possible. We’ve taken to thinking of Quantum Country as a memory laboratory. That is, it’s a system which can be used both to better understand how memory works, and also to develop new kinds of memory system. We’d like to answer questions such as:

    • What are new ways memory systems can be applied, beyond the simple, declarative knowledge of past systems?
    • How deep can the understanding developed through a memory system be? What patterns will help users deepen their understanding as much as possible?
    • How far can we raise the human capacity for memory? And with how much ease? What are the benefits and drawbacks?
    • Might it be that one day most human beings will have a regular memory practice, as part of their everyday lives? Can we make it so memory becomes a choice; is it possible to in some sense solve the problem of memory?
  533. https://blogs.nvidia.com/blog/2019/03/18/gaugan-photorealistic-landscapes-nvidia-research/

  534. http://nymag.com/news/intelligencer/modafinil-2013-4/

  535. https://www.nytimes.com/2014/02/23/health/a-hot-debate-over-e-cigarettes-as-a-path-to-tobacco-or-from-it.html

  536. https://www.obormot.net/demos/these-waifus-do-not-exist-v2-alt

  537. https://www.oglaf.com/glindr/

  538. ⁠, Oh You Pretty Things (2020):

    Filmmaker Gary Hustwit is streaming his documentaries free worldwide during the global COVID-19 crisis. Each Tuesday we’ll be posting another film here. We hope you enjoy them, and please stay strong.

    March 14 to 21: Helvetica
    March 24 to 31: Objectified
    March 31 to April 7: Urbanized
    April 7 to 14: Rams

    April 14 to 21: Workplace (2016, 64 minutes) is a film about the past, present, and future of the office…

    April 21 to 28: TBA
    April 28 to May 5: TBA

    [Oh You Pretty Things is a web shop run by a collective of filmmakers and visual artists based in Brooklyn NY. We make films, art, books, posters, photographs, clothing, and other fine stuff.]

  539. https://old.reddit.com/r/DarkNetMarkets/comments/1ue777/ask_lord_voldemort_anything/

  540. https://old.reddit.com/r/DataHoarder/

  541. https://old.reddit.com/r/IncreasinglyVerbose/

  542. https://old.reddit.com/r/MLPtunes/

  543. https://old.reddit.com/r/NavySealCopypasta/

  544. https://old.reddit.com/r/Piracy/comments/2oftbu/guide_the_idiot_proof_guide_to_downloading_ebooks/

  545. https://old.reddit.com/r/PrequelMemes/

  546. https://old.reddit.com/r/SilkRoad/comments/1tb2yl/sr_admin_and_mod_just_got_arrestedmy_boyfriend/

  547. https://old.reddit.com/r/TOUHOUMUSIC/search?q=author%3Agwern&sort=new&restrict_sr=on&t=all

  548. https://old.reddit.com/r/Vocaloid/search?q=author%3Agwern&sort=new&restrict_sr=on&t=all

  549. https://old.reddit.com/r/afinil/comments/3sgb9o/air_force_cadet_sentenced_to_3_years_after_buying/

  550. https://old.reddit.com/r/emojipasta/

  551. https://old.reddit.com/r/explorables/

  552. https://old.reddit.com/r/reinforcementlearning/search?q=flair%3AMeta-RL&sort=top&restrict_sr=on&t=all

  553. https://eva.onegeek.org/pipermail/evangelion/2006-September/003693.html

  554. https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1558-5646.1995.tb05954.x

  555. ⁠, Christine Payne (OpenAI) (2019-04-25):

    We’ve created MuseNet, a deep neural network that can generate 4-minute musical compositions with 10 different instruments, and can combine styles from country to Mozart to the Beatles. MuseNet was not explicitly programmed with our understanding of music, but instead discovered patterns of harmony, rhythm, and style by learning to predict the next token in hundreds of thousands of MIDI files. MuseNet uses the same general-purpose unsupervised technology as ⁠, a large-scale transformer model trained to predict the next token in a sequence, whether audio or text.

    [See also: ⁠, Child et al 2019

    Transformers are powerful sequence models, but require time and memory that grows quadratically with the sequence length. In this paper we introduce sparse factorizations of the attention matrix which reduce this to 𝒪(n ⋅ √n). We also introduce (a) a variation on architecture and initialization to train deeper networks, (b) the recomputation of attention matrices to save memory, and (c) fast attention kernels for training. We call networks with these changes Sparse ⁠, and show they can model sequences tens of thousands of timesteps long using hundreds of layers. We use the same architecture to model images, audio, and text from raw bytes, setting a new state of the art for density modeling of Enwik8, CIFAR-10, and ImageNet-64. We generate unconditional samples that demonstrate global coherence and great diversity, and show it is possible in principle to use self-attention to model sequences of length one million or more. ]

  556. ⁠, Anonymous (2020-09-28):

    We propose an efficient Transformer that eliminates attention.

    We introduce Attention Free Transformer (AFT), an efficient variant of Transformers that eliminates the need for spatial attention. AFT offers great simplicity compared with standard Transformers, where the multi-head attention operation is replaced with the composition of element-wise multiplications/​​​​divisions and global/​​​​local pooling. We provide several variants of AFT along with simple yet efficient implementations that are supported by main stream deep learning libraries. We show that, surprisingly, we are able to train AFT effectively on challenging benchmarks, and also to match or surpass the standard Transformer counterparts.

    [Keywords: Transformers, attention, efficient]

  557. https://osf.io/WX7Ck/

  558. https://orbis.stanford.edu/

  559. https://academic.oup.com/policing/advance-article/doi/10.1093/police/pay063/5094543

  560. https://ourworldindata.org/famines/

  561. https://www.overcomingbias.com/2012/02/dog-vs-cat-medicine.html

  562. http://aje.oxfordjournals.org/content/164/6/529.long

  563. ⁠, Xavier Marquez (2020-08-13; sociology  /​ ​​ ​preference-falsification):

    This chapter argues that leader personality cults are typically produced by a specific set of mechanisms of flattery inflation. It describes how loyalty signaling, emotional amplification, and direct production mechanisms can combine, under specific circumstances, to transform ordinary flattery into full-blown practices of ruler worship. And it argues for attending to the specific conditions that make possible the operation of these mechanisms, showing how patronage relationships in particular provide fertile ground for the emergence of personality cults. Moreover, the chapter argues that both ancient and modern leader cults depend on similar mechanisms, despite clear differences in context and function. I illustrate the operation of these mechanisms with many modern examples and an extended discussion of one ancient example, the abortive cult of Caligula during the Roman Principate.

    [Keywords: personality cults, Caligula, flattery inflation, Hugo Chávez, Mao Zedong, Stalin]

  564. ⁠, paperswithcode.com (2019-08-28):

    Language modeling is the task of predicting the next word or character in a document. This page lists key recent papers on NLP language modeling and records reported research performance on the following tasks: WikiText-103, Penn Treebank (Word Level), enwiki8, Text8, One Billion Word, WikiText-2, Hutter Prize, Penn Treebank (Character Level).

  565. ⁠, Omar Shehata (2019-05-01):

    [‘Explorable’ web page on the venerable JPEG format. Shehata provides an interactive interface to raw JPEG files: one can edit the raw hex files of a JPEG to instantly see the effects of the corruption on the displayed image. By editing rows, one understands the redundancy of the pixels in an image, and how the JPEG format exploits this by building up images out of a series of repeating blocks, then achieving its size reductions by throwing away the least important blocks.]

  566. http://lacey.patch.com/articles/local-boys-make-good-in-tinseltown-with-immortals-75674a48

  567. https://www.patreon.com/posts/39864473

  568. https://patrickcollison.com/questions

  569. http://paulgraham.com/useful.html

  570. http://pcdb.santafe.edu/

  571. https://pcpartpicker.com/list/nwQnZ8

  572. ⁠, Richard Wiseman, Caroline Watt, Diana Kornbrot (2019-01-16):

    The recent ‘replication crisis’ in psychology has focused attention on ways of increasing methodological rigor within the behavioral sciences. Part of this work has involved promoting ‘Registered Reports’, wherein journals peer review papers prior to data collection and publication. Although this approach is usually seen as a relatively recent development, we note that a prototype of this publishing model was initiated in the mid-1970s by parapsychologist Martin Johnson in the European Journal of Parapsychology (EJP). A retrospective and observational comparison of Registered and non- published in the EJP during a seventeen-year period provides circumstantial evidence to suggest that the approach helped to reduce questionable research practices. This paper aims both to bring Johnson’s pioneering work to a wider audience, and to investigate the positive role that Registered Reports may play in helping to promote higher methodological and statistical standards.

    …The final dataset contained 60 papers: 25 RRs and 35 non-RRs. The RRs described 31 experiments that tested 131 hypotheses, and the non-RRs described 60 experiments that tested 232 hypotheses.

    28.4% of the statistical tests reported in non-RRs were statistically-significant (66⁄232: 95% [21.5%–36.4%]); compared to 8.4% of those in the RRs (11⁄131: 95% CI [4.0%–16.8%]). A simple 2 × 2 contingency analysis showed that this difference is highly statistically-significant (Fisher’s exact test: p < 0.0005, Pearson chi-square = 20.1, Cohen’s d = 0.48).

    …Parapsychologists investigate the possible existence of phenomena that, for many, have a low a priori likelihood of being genuine (see, eg., Wagenmakers et al 2011). This has often resulted in their work being subjected to a considerable amount of critical attention (from both within and outwith the field) that has led to them pioneering several methodological advances prior to their use within mainstream psychology, including the development of randomisation in experimental design (Hacking, 1988), the use of blinds (Kaptchuk, 1998), explorations into randomisation and statistical inference (Fisher, 1924), advances in replication issues (Rosenthal, 1986), the need for pre-specification in meta-analysis (Akers, 1985; Milton, 1999; Kennedy, 2004), and the creation of a formal study registry (Watt, 2012; Watt & Kennedy, 2015). Johnson’s work on RRs provides another striking illustration of this principle at work.

  573. http://perma.cc/

  574. http://test.personality-project.org/

  575. http://www.personalitytest.net/ipip/index.html

  576. http://philanthropy.com/article/FinancialLeadership-Woes/138335/

  577. https://philarchive.org/archive/TURMAA-6v2

  578. https://philpapers.org/surveys/results.pl

  579. https://pinboard.in/

  580. https://plato.stanford.edu/entries/epistemic-paradoxes/#MooPro

  581. ⁠, Daniel Smilkov, Shan Carter (2016-11-05):

    [(Github) An in-browser implementation using TypeScript of a simple feedforward MLP, whose architecture, LR, activation, regularization, and task can be varied and the NN retrained with the intermediate function of each neuron visualized and the decision boundary on the data plotted. One can see how different hyperparameters lead to different learned units and boundaries of varying smoothness and shapes, and how updates it each iteration.

    Available settings:

    • Show test data
    • Discretize output
    • Play button
    • Step button
    • Reset button
    • Learning rate
    • Activation
    • Regularization
    • Regularization rate
    • Problem type
    • Which dataset
    • Ratio train data
    • Noise level
    • Batch size
    • number of hidden layers]
  582. ⁠, John R. Shaffer, Ekaterina Orlova, Myoung Keun Lee, Elizabeth J. Leslie, Zachary D. Raffensperger, Carrie L. Heike, Michael L. Cunningham, Jacqueline T. Hecht, Chung How Kau, Nichole L. Nidey, Lina M. Moreno, George L. Wehby, Jeffrey C. Murray, Cecelia A. Laurie, Cathy C. Laurie, Joanne Cole, Tracey Ferrara, Stephanie Santorico, Ophir Klein, Washington Mio, Eleanor Feingold, Benedikt Hallgrimsson, Richard A. Spritz, Mary L. Marazita, Seth M. Weinberg (2016-06-08):

    Numerous lines of evidence point to a genetic basis for facial morphology in humans, yet little is known about how specific genetic variants relate to the phenotypic expression of many common facial features. We conducted genome-wide association meta-analyses of 20 quantitative facial measurements derived from the 3D surface images of 3118 healthy individuals of European ancestry belonging to two US cohorts. Analyses were performed on just under one million genotyped SNPs (Illumina OmniExpress+Exome v1.2 array) imputed to the 1000 Genomes reference panel (Phase 3). We observed genome-wide statistically-significant associations (p < 5 x 10−8) for cranial base width at 14q21.1 and 20q12, intercanthal width at 1p13.3 and Xq13.2, nasal width at 20p11.22, nasal ala length at 14q11.2, and upper facial depth at 11q22.1. Several genes in the associated regions are known to play roles in craniofacial development or in syndromes affecting the face: MAFB, PAX9, MIPOL1, ALX3, HDAC8, and PAX1. We also tested genotype-phenotype associations reported in two previous genome-wide studies and found evidence of replication for nasal ala length and SNPs in CACNA2D3 and PRDM16. These results provide further evidence that common variants in regions harboring genes of known craniofacial function contribute to normal variation in human facial features. Improved understanding of the genes associated with facial morphology in healthy individuals can provide insights into the pathways and mechanisms controlling normal and abnormal facial morphogenesis.

    Author Summary:

    There is a great deal of evidence that genes influence facial appearance. This is perhaps most apparent when we look at our own families, since we are more likely to share facial features in common with our close relatives than with unrelated individuals. Nevertheless, little is known about how variation in specific regions of the genome relates to the kinds of distinguishing facial characteristics that give us our unique identities, e.g., the size and shape of our nose or how far apart our eyes are spaced. In this paper, we investigate this question by examining the association between genetic variants across the whole genome and a set of measurements designed to capture key aspects of facial form. We found evidence of genetic associations involving measures of eye, nose, and facial breadth. In several cases, implicated regions contained genes known to play roles in embryonic face formation or in syndromes in which the face is affected. Our ability to connect specific genetic variants to ubiquitous facial traits can inform our understanding of normal and abnormal craniofacial development, provide potential predictive models of evolutionary changes in human facial features, and improve our ability to create forensic facial reconstructions from DNA.

  583. ⁠, Heleen A. Slagter, Antoine Lutz, Lawrence L. Greischar, Andrew D. Francis, Sander Nieuwenhuis, James M. Davis, Richard J. Davidson (2007-03-14):

    The information processing capacity of the human mind is limited, as is evidenced by the so-called “attentional-blink” deficit: When two targets (T1 and T2) embedded in a rapid stream of events are presented in close temporal proximity, the second target is often not seen. This deficit is believed to result from competition between the two targets for limited attentional resources. Here we show, using performance in an attentional-blink task and scalp-recorded brain potentials, that meditation, or mental training, affects the distribution of limited brain resources. Three months of intensive mental training resulted in a smaller attentional blink and reduced brain-resource allocation to the first target, as reflected by a smaller T1-elicited P3b, a brain-potential index of resource allocation. Furthermore, those individuals that showed the largest decrease in brain-resource allocation to T1 generally showed the greatest reduction in attentional-blink size. These observations provide novel support for the view that the ability to accurately identify T2 depends upon the efficient deployment of resources to T1. The results also demonstrate that mental training can result in increased control over the distribution of limited brain resources. Our study supports the idea that plasticity in brain and mental function exists throughout life and illustrates the usefulness of systematic mental training in the study of the human mind.

    Author Summary:

    Meditation includes the mental training of attention, which involves the selection of goal-relevant information from the array of inputs that bombard our sensory systems. One of the major limitations of the attentional system concerns the ability to process two temporally close, task-relevant stimuli. When the second of two target stimuli is presented within a half second of the first one in a rapid sequence of events, it is often not detected. This so-called “attentional-blink” deficit is thought to result from competition between stimuli for limited attentional resources. We measured the effects of intense meditation on performance and scalp-recorded brain potentials in an attentional-blink task. We found that three months of intensive meditation reduced brain-resource allocation to the first target, enabling practitioners to more often detect the second target with no compromise in their ability to detect the first target. These findings demonstrate that meditative training can improve performance on a novel task that requires the trained attentional abilities.

    Intensive training in Vipassana meditation enhances one’s ability to allocate attention efficiently in order to detect visual targets accurately. Behavioral and event-related potential evidence for a causal link between behavioral training and brain plasticity in adults is shown.

  584. ⁠, Peter M. Visscher, Sarah E. Medland, Manuel A. R. Ferreira, Katherine I. Morley, Gu Zhu, Belinda K. Cornes, Grant W. Montgomery, Nicholas G. Martin (2006-02-06):

    The study of continuously varying, quantitative traits is important in evolutionary biology, agriculture, and medicine. Variation in such traits is attributable to many, possibly interacting, genes whose expression may be sensitive to the environment, which makes their dissection into underlying causative factors difficult. An important population parameter for quantitative traits is heritability, the proportion of total variance that is due to genetic factors. Response to artificial and and the degree of resemblance between relatives are all a function of this parameter. Following the classic paper by in 1918, the estimation of additive and dominance genetic and heritability in populations is based upon the expected proportion of genes shared between different types of relatives, and explicit, often controversial and untestable models of genetic and non-genetic causes of family resemblance. With genome-wide coverage of genetic markers it is now possible to estimate such parameters solely within families using the actual degree of identity-by-descent sharing between relatives. Using genome scans on 4,401 quasi-independent sib pairs of which 3,375 pairs had phenotypes, we estimated the heritability of height from empirical genome-wide identity-by-descent sharing, which varied from 0.374 to 0.617 (mean 0.498, standard deviation 0.036). The variance in identity-by-descent sharing per chromosome and per genome was consistent with theory. The maximum likelihood estimate of the heritability for height was 0.80 with no evidence for non-genetic causes of sib resemblance, consistent with results from independent twin and family studies but using an entirely separate source of information. Our application shows that it is feasible to estimate genetic variance solely from within-family segregation and provides an independent validation of previously untestable assumptions. Given sufficient data, our new paradigm will allow the estimation of genetic variation for disease susceptibility and quantitative traits that is free from confounding with non-genetic factors and will allow partitioning of genetic variation into additive and non-additive components.


    Quantitative geneticists attempt to understand variation between individuals within a population for traits such as height in humans and the number of bristles in fruit flies. This has been traditionally done by partitioning the variation in underlying sources due to genetic and environmental factors, using the observed amount of variation between and within families. A problem with this approach is that one can never be sure that the estimates are correct, because nature and nurture can be confounded without one knowing it. The authors got around this problem by comparing the similarity between relatives as a function of the exact proportion of genes that they have in common, looking only within families. Using this approach, the authors estimated the amount of total variation for height in humans that is due to genetic factors from 3,375 sibling pairs. For each pair, the authors estimated the proportion of genes that they share from DNA markers. It was found that about 80% of the total variation can be explained by genetic factors, close to results that are obtained from classical studies. This study provides the first validation of an estimate of genetic variation by using a source of information that is free from nature–nurture assumptions.

  585. ⁠, Alessandro Liberati, Douglas G. Altman, Jennifer Tetzlaff, Cynthia Mulrow, Peter C. Gøtzsche, John P. A. Ioannidis, Mike Clarke, P. J. Devereaux, Jos Kleijnen, David Moher ():

    Alessandro Liberati and colleagues present an Explanation and Elaboration of the PRISMA Statement, updated guidelines for the reporting of systematic reviews and meta-analyses.

    Systematic reviews and meta-analyses are essential to summarize evidence relating to efficacy and safety of health care interventions accurately and reliably. The clarity and transparency of these reports, however, is not optimal. Poor reporting of systematic reviews diminishes their value to clinicians, policy makers, and other users.

    Since the development of the QUOROM (QUality Of Reporting Of Meta-analysis) Statement—a reporting guideline published in 1999—there have been several conceptual, methodological, and practical advances regarding the conduct and reporting of systematic reviews and meta-analyses. Also, reviews of published systematic reviews have found that key information about these studies is often poorly reported. Realizing these issues, an international group that included experienced authors and methodologists developed PRISMA (Preferred Reporting Items for Systematic reviews and Meta-Analyses) as an evolution of the original QUOROM guideline for systematic reviews and meta-analyses of evaluations of health care interventions.

    The PRISMA Statement consists of a 27-item checklist and a four-phase flow diagram. The checklist includes items deemed essential for transparent reporting of a systematic review. In this Explanation and Elaboration document, we explain the meaning and rationale for each checklist item. For each item, we include an example of good reporting and, where possible, references to relevant empirical studies and methodological literature. The PRISMA Statement, this document, and the associated Web site (http:    /​ ​​ ​​ ​​ ​    /​ ​​ ​​ ​​ ​www.prisma-statement.org    /​ ​​ ​​ ​​ ​) should be helpful resources to improve reporting of systematic reviews and meta-analyses.

  586. ⁠, Chih-Da Wu, Eileen McNeely, J. G. Cedeño-Laurent, Wen-Chi Pan, Gary Adamkiewicz, Francesca Dominici, Shih-Chun Candice Lung, Huey-Jen Su, John D. Spengler (2014-08-29):

    Various studies have reported the physical and mental health benefits from exposure to “green” neighborhoods, such as proximity to neighborhoods with trees and vegetation. However, no studies have explicitly assessed the association between exposure to “green” surroundings and cognitive function in terms of student academic performance. This study investigated the association between the “greenness” of the area surrounding a Massachusetts public elementary school and the academic achievement of the school’s student body based on standardized tests with an ecological setting. Researchers used the composite school-based performance scores generated by the Massachusetts Comprehensive Assessment System (MCAS) to measure the percentage of 3rd-grade students (the first year of standardized testing for 8–9 years-old children in public school), who scored “Above Proficient” (AP) in English and Mathematics tests (Note: Individual student scores are not publically available). The MCAS results are comparable year to year thanks to an equating process. Researchers included test results from 2006 through 2012 in 905 public schools and adjusted for differences between schools in the final analysis according to race, gender, English as a second language (proxy for ethnicity and language facility), parent income, student-teacher ratio, and school attendance. Surrounding greenness of each school was measured using satellite images converted into the Normalized Difference Vegetation Index (NDVI) in March, July and October of each year according to a 250-meter, 500-meter, 1,000-meter, and 2000-meter circular buffer around each school. Spatial Generalized Linear Mixed Models (GLMMs) estimated the impacts of surrounding greenness on school-based performance. Overall the study results supported a relationship between the “greenness” of the school area and the school-wide academic performance. Interestingly, the results showed a consistently positive statistically-significant association between the greenness of the school in the Spring (when most Massachusetts students take the MCAS tests) and school-wide performance on both English and Math tests, even after adjustment for socio-economic factors and urban residency.

  587. https://plus.google.com/u/0/103530621949492999968/posts/fqxuM2SBRQ5

  588. https://www.pnas.org/content/early/2020/03/17/1915769117

  589. https://www.poetryfoundation.org/poems-and-poets/poems/detail/44400

  590. https://www.codespaces.com/power-searching-with-google.html

  591. https://practicaltypography.com/alternate-figures.html#oldstyle-figures

  592. https://predictionbook.com/predictions/3623

  593. http://press.etc.cmu.edu/index.php/product/well-played-3-0/

  594. ⁠, Progress Studies for Young Scholars (2020-05):

    [2020] Progress Studies for Young Scholars is an online program of guided self-study in the history of industrial civilization for high school students.

    This program will explore: what problems, challenges and hardships in life and work were faced by people in earlier generations and centuries? And how did we solve those problems through science, technology, and invention?

    Learn about manufacturing from blacksmiths to assembly lines; about power from water wheels to combustion to electricity; about food from famine to industrial agriculture and genetically modified crops; about disease from basic sanitation to scientific medicine—and the struggles and circumstances of the men and women who worked to bend the arc of humanity upward.

    Your learning will be supported by instructors who will help you develop your reasoning and research skills. You’ll also have the chance to engage ideas with a community of like-minded peers…A six-week course with daily reading, audio or video content. Go through it on your own, or join a study group with an instructor for daily online discussions and Q&A.

    Speaker series: Danica Remy · Deirdre Nansen McCloskey · Adam Mossoff · Anton Howes · Joel Mokyr · Laura Mazer · Manjari Narayan · Matt Bateman · Max Roser · Sarah Constantin · Tyler Cowen · Jason Crawford · Jerry Neumann · Michael Dearing · Michael Strong · Noor Siddiqui · Patrick Collison · Samo Burja

  595. http://projecteuclid.org/DPubS/Repository/1.0/Disseminate?view=body&id=pdfview_1&handle=euclid.ssu/1356628931

  596. http://www.projectrho.com/public_html/rocket/spacegunexotic.php#propulsion

  597. https://www.proofofexistence.com

  598. http://www.proquest.com/en-US/products/dissertations/

  599. https://psyarxiv.com/gjh95/

  600. http://pro.psychcentral.com/sparlon-and-adhd-the-power-of-a-7-year-old/002889.html

  601. http://www.psychiatryinvestigation.org/journal/view.php?number=865

  602. http://www.psychservices.psychiatryonline.org/data/Journals/AJP/3393/811.pdf

  603. http://www.psychologytoday.com/files/attachments/56143/wai-americas-elite-2013.pdf

  604. https://qntm.org/urls

  605. https://quantum.country/qcvc

  606. https://www.quora.com/What-do-biologists-think-of-gwerns-criticism-of-Folding-Home

  607. https://qwantz.com/index.php?comic=2366

  608. https://www.r-bloggers.com/visualization-series-using-scatterplots-and-models-to-understand-the-diamond-market-so-you-dont-get-ripped-off/

  609. http://www.r-inla.org/

  610. https://racket-lang.org/

  611. https://www.rand.org/pubs/research_reports/RR1607.html

  612. https://rationality.org/

  613. http://bjp.rcpsych.org/content/181/1/22.long

  614. https://www.rdocumentation.org/packages/EnvStats/versions/2.1.0/topics/evNormOrdStats

  615. https://readonlymemory.vg/shop/book/arcade-game-typography/

  616. http://book.realworldhaskell.org/read/data-structures.html#id637702

  617. http://archive.recapthelaw.org/search/advanced/?q=modafinil

  618. https://old.reddit.com/r/DarkNetMarkets/comments/2u26ij/uncategorized_i_bought_100_superman_pills_from/

  619. https://old.reddit.com/user/gwern/submitted/

  620. https://www.researchgate.net/publication/232606103_Cramming_A_Barrier_to_Student_Success_a_Way_to_Beat_the_System_or_an_Effective_Learning_Strategy

  621. https://www.ribbonfarm.com/2010/07/26/a-big-little-idea-called-legibility/

  622. https://ricon.dev/

  623. https://www.rifters.com/crawl/?p=6116

  624. http://www.rocketpunk-manifesto.com/2009/06/space-warfare-i-gravity-well.html

  625. ⁠, Jason Crawford (2020-01-28):

    [Public health history review: as famous as vaccines and antibiotics were, deaths from infectious diseases had been declining for centuries before hand, and vaccines/​​​​antibiotics merely helped continue the decline without representing a major trend break. Such trends date back to long before the vindication of germ theory, as incorrect theories like miasmas nevertheless led to effective sanitation and cleanliness interventions which did reduce disease: “the mortality data points to a large and easy-to-underappreciate role of pest control, water sanitation, food handling, and general hygiene.”]

  626. https://row1.ca/pixels-and-their-neighbors

  627. http://rstb.royalsocietypublishing.org/content/365/1537/73.full

  628. http://rpubs.com/EmilOWK/PING_admixture_study

  629. https://safebooru.org/index.php?page=post&s=list&tags=heterochromia

  630. http://journals.sagepub.com/doi/abs/10.1177/1948550617707015

  631. https://www.schneier.com/blog/archives/2008/03/the_security_mi_1.html

  632. http://scholar.google.com/scholar?q=noopept

  633. http://www.scholarpedia.org/article/N-body_choreographies

  634. ⁠, Tanner Greer (2015-09-30):

    …A good place to start is with the Webster-Hayne debate of 1830. Of all American oratory, only the Lincoln-Douglass debates can claim greater fame than the debate Daniel Webster and Robert Hayne held on the antebellum Senate floor. At that time there was a resolution before the Senate calling for all new federal land surveys to be postponed until all of the existing land already surveyed was sold. This struck the ire of the westerners, who pushed for federal land to be given to new settlers without charge or delay…These allusions to Shakespeare only occupy a normal portion of the two men’s debate—no more than a few paragraphs out of ninety or so pages of text. Nevertheless, the use of Macbeth’s script in the debate is telling. Neither Webster nor Hayne thought it was a waste of their time to debate the finer points of Shakespeare’s plays in the halls of the Senate. The reader senses that Webster, in particular, did so in a positively gleeful fashion.

    What has happened here? How have we gone from long discussions of Shakespearean drama on the senate floor to the shallow repetition of disembodied sentence fragments? The answers to this question tell us much about the American body politic:

    1. The decline of public speaking as a vital part of American culture. Oratory is something of a lost art in modern America. It is hard to imagine just how vital it was to public life for most of America’s history. In Webster’s day public speaking was a central part of entertainment, education, civic life, and religious practice. He was elected in the midst of the 2nd Great Awakening, when American religious life was dominated by camp meetings and church members were expected to preach and testify one to another. It was a time when every township had a lyceum at its center, and intellectual life was dominated by those who traveled the lyceum circuit. Collections of speeches like The Columbian Orator were the most common type of schoolbook in the antebellum era, while most American men actively participated in town assemblies and party caucuses. The mastery of proper political rhetoric was an essential social skill.

    Add all this together and you are left with a population that found immense pleasure in listening to, reenacting, and reading the speeches of others. It was a prized art, and when masters like Webster or Lincoln displayed their talents, people flocked together to listen to them. There was thus a great deal of patience for the sort of rhetorical flourish inherit in the long discussions of Shakespeare seen above. Today’s Americans will not sit still and listen to a political speech for longer than ten minutes. The medium through which politicians communicate to the masses really doesn’t let them. Radio shows and news channels rely on the soundbite. If a politician’s message cannot be squeezed into a seven second slot it will not be heard.

  635. http://www.sciencebasedmedicine.org/index.php/antioxidants-and-exercise-more-harm-than-good/

  636. https://scienceblogs.com/insolence/2009/07/06/popularity-versus-reliability-in-medical

  637. http://www.sciencedaily.com/releases/2007/05/070525204143.htm

  638. https://www.sciencedirect.com/science/article/pii/S0160289618300266

  639. https://blogs.sciencemag.org/pipeline/archives/2013/11/26/of_mice_studies_and_men

  640. http://www.sciencenews.org/view/generic/id/65600/title/Fish_oil_fails_to_hold_off_heart_arrhythmia

  641. http://www.sciencenewsline.com/medicine/2012010922310082.html

  642. http://www.sciencepubs.org/content/345/6200/1074.full.pdf

  643. http://www.scientificamerican.com/article/china-s-bold-push-into-genetically-customized-animals/

  644. https://conference.scipy.org/proceedings/scipy2013/pdfs/bergstra_hyperopt.pdf

  645. https://physicstoday.scitation.org/do/10.1063/PT.6.1.20180822a/full/

  646. https://scp-wiki.wikidot.com/antimemetics-division-hub#toc1

  647. https://pdfs.semanticscholar.org/f116/27888a7e0b1cb7077dff91c9d97df1cad062.pdf

  648. http://www.sendspace.com/file/ozr19p

  649. http://serendipityrecs.com/

  650. http://www.blog.sethroberts.net/category/nutrition/omega-3/omega-3-directory/

  651. https://www.sevensecularsermons.org/about/

  652. https://www.shawwn.com/swarm

  653. https://gwern.shinyapps.io/orderStatisticsIncreasedVariance/

  654. http://sifter.org/iqtest/

  655. https://www.simplify.so/

  656. http://interviews.slashdot.org/story/11/09/06/1458254/Kevin-Kelly-Answers-Your-Questions

  657. https://slate.com/news-and-politics/2013/08/israel-keyes-serial-killer-the-most-meticulous-serial-killer-of-modern-times-is-taunting-the-fbi-from-beyond-the-grave.html

  658. https://slatestarcodex.com/2014/09/24/book-review-red-plenty/

  659. https://www.smbc-comics.com/

  660. http://www.smithsonianmag.com/history/crockfords-club-how-a-fishmonger-built-a-gambling-hall-and-bankrupted-the-british-aristocracy-148268691/

  661. https://www.snopes.com/fact-check/school-lemonade-laxatives/

  662. http://www.snpedia.com/index.php/Genomes

  663. https://soundcloud.com/nago-music/nago-blue-momentary

  664. http://pgpry.sourceforge.net/

  665. http://www.sparkfun.com/commerce/tutorials.php

  666. https://spreadsheets.google.com/ccc?key=rOfijmsJ-hxPbzmbi4dmHVg

  667. https://link.springer.com/article/10.1007/s11292-017-9297-z

  668. https://media.springernature.com/full/springer-static/image/art%3A10.1186%2Fs12917-017-0987-6/MediaObjects/12917_2017_987_Fig7_HTML.gif

  669. https://www.thessgac.org/data

  670. http://crypto.stackexchange.com/questions/5831/what-is-the-progress-on-the-mit-lcs35-time-capsule-crypto-puzzle

  671. https://stackroboflow.com/

  672. http://research.stlouisfed.org/fred2/series/JPNURYNAA

  673. https://www.straighttalkonevidence.org/2018/10/11/large-randomized-trial-finds-disappointing-effects-for-federally-funded-programs-to-facilitate-the-re-entry-of-prisoners-into-the-community-a-new-approach-is-needed/

  674. https://streamable.com/87z73

  675. http://stylegan.xyz/video

  676. https://taibbi.substack.com/p/the-news-media-is-destroying-itself

  677. ⁠, Adam King (2019):

    [Interactive web interface to -1.5b.] See how a modern neural network completes your text. Type a custom snippet or try one of the examples…Built by Adam King (@AdamDanielKing) as an easier way to play with OpenAI’s new machine learning model. This site runs the full-sized GPT-2 model, called 1558M.

  678. http://www.tandfonline.com/doi/pdf/10.1080/01459741003715391

  679. http://www.tarsnap.com/scrypt.html

  680. http://tasvideos.org/4156S.html

  681. https://www.ted.com/talks/christopher_ryan_are_we_designed_to_be_sexual_omnivores

  682. https://www.theatlantic.com/magazine/archive/2005/05/the-coming-death-shortage/4105/

  683. https://thecleverest.com/judgefakepeople/main.php?sort=highest

  684. https://thefirstaibook.com/#post-inner

  685. https://thegradient.pub/understanding-evaluation-metrics-for-language-models/

  686. https://www.theparisreview.org/interviews/6012/ray-bradbury-the-art-of-fiction-no-203-ray-bradbury

  687. http://thepharmacyexpress.com/Products2.asp?Brand=Modvigil+%28Provigil%2C+Modalert%2C+Modapro%2C+Generic+Modafinil%29&T=d

  688. https://thepiratebay.org/torrent/14045031/My_Little_Pony_Friendship_is_Magic_Season_1-5

  689. http://thesecatsdonotexist.com/

  690. https://web.archive.org/web/20091214014113/http://theuncertainfuture.com/faq.html#7

  691. https://www.theverge.com/2015/5/13/8592817/rat-patrol-new-york-alberta-canada-south-georgia-eradication

  692. https://thisanimedoesnotexist.ai/downloads.html

  693. https://thiscardoesnotexist.glitch.me/

  694. https://thiscatdoesnotexist.com/

  695. https://thiseyedoesnotexist.com/story/

  696. ⁠, Arfafax (2020-05-07):

    A showcase: high-quality GAN-generated furry (anthropomorphic animals) faces, trained on cropped from the e621 furry image booru. For higher quality, the creator heavily filtered faces and aligned them, and upscaled using waifu2×. For display, it reuses Obormot’s “These Waifus Do Not Exist” scrolling grid code to display an indefinite number of faces rather than one at a time. (TFDNE is also available on Artbreeder for interactive editing/​​​​crossbreeding, and a for Ganspace-based editing.)

    9 random TFDNE furry face samples in a grid

    Model download mirrors:

    • Google Drive

    • Mega

    • Rsync:

      rsync --verbose rsync:// ./

    Previously, ⁠; later: the My Little Pony-themed followup ⁠, and ⁠.

  697. http://thismarketingblogdoesnotexist.com/

  698. ⁠, Phillip Wang (2019-02-12):

    [This Person Does Not Exist is a StyleGAN-based noninteractive website, which uses the Nvidia-trained FFHQ face StyleGAN model to generate realistic 1024px faces (upgraded to in 2020). One face is displayed at a time, and a new face automatically generated every few seconds. It inspired a rash of copycats, including ⁠.]

    Recently a talented group of researchers at Nvidia released the current state of the art generative adversarial network, StyleGAN, over at https:/​​​​/​​​​github.com/​​​​NVlabs/​​​​stylegan I have decided to dig into my own pockets and raise some public awareness for this technology. Faces are most salient to our cognition, so I’ve decided to put that specific pretrained model up. Their research group have also included pretrained models for cats, cars, and bedrooms in their repository that you can immediately use. Each time you refresh the site, the network will generate a new facial image from scratch from a 512 dimensional vector.

  699. ⁠, Arfafax (2020-07):

    “This Pony Does Not Exist” (TPDNE) is the followup to ⁠, also by Arfafax. He scraped the Derpibooru My Little Pony: Friendship is Magic image booru, hand-annotated images and trained a pony face YOLOv3 cropper to create a pony face crop dataset⁠, and trained the TFDNE StyleGAN 2 model to convergence on TensorFork TPU pods, with an upgrade to 1024px resolution via transfer learning/​​​​model surgery. The interface reuses Said Achmiz’s These Waifus Do Not Exist grid UI.

    10 random pony samples from TPDNE; see also Derpibooru uploads from TPDNE.

    The S2 model snapshot is available for download and I have mirrored it (rsync rsync:// ./). See also: ⁠/​​​​

  700. http://thisrentaldoesnotexist.com/

  701. https://www.thisstorydoesnotexist.com/

  702. https://thisvesseldoesnotexist.com/

  703. ⁠, Gwern Branwen (2019-02-19):

    (TWDNE) is a static website which uses JS to display random anime faces generated by StyleGAN neural networks, along with GPT-3-generated anime plot summaries. Followups: ⁠/​​​​⁠/​​​​⁠.

    A screenshot of “This Waifu Does Not Exist” (TWDNE) showing a random StyleGAN-generated anime face and a random GPT-3 text sample conditioned on anime keywords/​​​​phrases.
  704. http://www.timeanddate.com/date/durationresult.html?m1=9&d1=16&y1=2012&m2=3&d2=9&y2=2013&ti=on

  705. https://tineye.com/

  706. https://tl.net/blogs/319375-bronze-part-3-casually-cruel?view=all

  707. http://tom7.org/mario/

  708. https://www.tomshardware.com/news/google-removing-minix-management-engine-intel,35876.html

  709. https://toolbox.google.com/datasetsearch/search?query=poem%20OR%20poetry&docid=T3haTlmLU9Dl6xqYAAAAAA%3D%3D

  710. http://tools.wmflabs.org/xtools/pages/index.php?name=Marudubshinki&lang=en&wiki=wikipedia&namespace=0&redirects=noredirects

  711. https://5onwnspjvuk7cwvk.tor2web.org/

  712. http://www.toranoana.jp/bl/article/04/0030/04/86/040030048682.html

  713. https://www.torproject.org/docs/installguide.html.en

  714. https://www.torservers.net/donate.html

  715. http://en.touhouwiki.net/wiki/Doujin_circles

  716. https://translate.google.com/translate?sl=auto&tl=en&js=y&prev=_t&hl=en&ie=UTF-8&u=http%3A%2F%2Fwww.sydsvenskan.se%2F2016-07-15%2Fatal-for-storskalig-narkotikahandel&edit-text=&act=url

  717. https://www.treasurydirect.gov/govt/reports/pd/histdebt/histdebt_histo5.htm

  718. https://trends.google.com/trends/explore?q=Long%20Bets

  719. http://www.tryhaskell.org

  720. https://tvtropes.org/pmwiki/pmwiki.php/Main/TheRenaissanceAgeOfAnimation

  721. https://www.twitch.tv/videos/416276005#openai

  722. http://twitter.com/intent/user?screen_name=Hiramatz&tw_i=303521521249447936

  723. https://popcon.ubuntu.com/

  724. https://unesdoc.unesco.org/ark:/48223/pf0000220391

  725. https://unsongbook.com/interlude-%D7%A9-obama/

  726. http://www.uptontea.com/shopcart/item.asp?itemID=BA20

  727. https://www.urbandictionary.com/define.php?term=Snape%20killed%20Dumbledore

  728. http://www.urth.net/urth/archives/v0010/0111.shtml

  729. https://www.uscourts.gov/services-forms/federal-court-reporting-program

  730. https://www.usenix.org/system/files/conference/usenixsecurity18/sec18-yuan_0.pdf

  731. https://usesthis.com/

  732. https://www.vanityfair.com/news/business/2014/04/g4s-global-security-company

  733. https://vast.ai/

  734. https://www.vice.com/en_us/article/59755n/the-speedrunner-who-wasnt-how-a-community-dealt-with-an-elaborate-cheater

  735. http://videolectures.net/rldm2015_silver_reinforcement_learning/

  736. https://vimeo.com/299232829/89a3ff9ae4

  737. https://vision-explorer.allenai.org/text_to_image_generation

  738. https://vizhub.healthdata.org/gbd-compare/#settings=1c5235ed0e4875665b1e2005644f6a7694b65e96

  739. http://www.vox.com/2015/2/4/7972077/against-oversleeping

  740. https://www.w3.org/International/wiki/Case_folding

  741. http://waifu2x.udp.jp/

  742. ⁠, Sizigi Studios (2019-07-23):

    [Waifu Labs is an interactive website for generating (1024px?) anime faces using a customized StyleGAN trained on Danbooru2018. Similar to Artbreeder, it supports face exploration and face editing, and at the end, a user can purchase prints of a particular face.]

    We taught a world-class artificial intelligence how to draw anime. All the drawings you see were made by a non-human artist! Wild, right? It turns out machines love waifus almost as much as humans do. We proudly present the next chapter of human history: lit waifu commissions from the world’s smartest AI artist. In less than 5 minutes, the artist learns your preferences to make the perfect waifu just for you.

  743. https://www.washingtonpost.com/wp-dyn/content/article/2006/03/23/AR2006032301459.html

  744. http://replay.waybackmachine.org/20071201121817/http://code.google.com/soc/2007/haskell/appinfo.html?csaid=637BFC2B6B13D512

  745. https://webrecorder.io/

  746. https://websitedownloader.io

  747. https://wellcomelibrary.org/item/b18032217

  748. http://www.whichfaceisreal.com/

  749. http://samuraijack.fandom.com/wiki/Episode_XL:_Jack_vs._the_Ninja

  750. http://en.wikibooks.org/wiki/User:Gwern

  751. https://en.wikichip.org/wiki/intel/core_i9/i9-7900x

  752. https://en.wikifur.com/wiki/History

  753. https://meta.wikimedia.org/wiki/InternetArchiveBot

  754. https://en.wikiquote.org/wiki/Randall_Jarrell

  755. https://en.wikisource.org/wiki/On_Authorship

  756. http://www.wikiwix.com/

  757. https://en.wiktionary.org/wiki/e%27en

  758. ⁠, Ciro Della Monica, Giuseppe Atzori, Derk-Jan Dijk (2015-06-12):

    Recently, evidence has emerged that the phases of the moon may modulate subjective sleep quality and polysomnographically assessed sleep structure in humans. We aimed to explore further the putative effects of circa-lunar periodicity (~29.5 days) on subjective and objective parameters of human sleep in a retrospective analysis. The baseline sleep recordings of 205 (91 males and 114 females; mean age = 47.47 years, standard deviation = 19.01; range: 20–84 years) healthy and carefully screened participants who participated in two clinical trials in the Surrey Clinical Research Centre were included in the analyses. Sleep was recorded in windowless sleep laboratories. For each study night, we calculated the distance, in days, to the date of the closest full moon phase and based on this distance, classified sleep records in three lunar classes. Univariate analysis of variance with factors lunar class, age and sex was applied to each of 21 sleep parameters. No statistically-significant main effect for the factor lunar class was observed for any of the objective sleep parameters and subjective sleep quality but some statistically-significant interactions were observed. The interaction between lunar class and sex was statistically-significant for total sleep time, Stage 4 sleep and rapid eye movement (REM) sleep. Separate analyses for men and women indicated that in women total sleep time, Stage 4 sleep and REM sleep were reduced when sleep occurred close to full moon, whereas in men REM duration increased around full moon. These data provide limited evidence for an effect of lunar phase on human sleep.

  759. http://www.winehq.org/pipermail/wine-devel/2002-February/003912.html

  760. http://www.wolfewiki.com/pmwiki/pmwiki.php?n=WolfeWiki.Introduction

  761. http://www.wolframalpha.com/input/?i=binomial+distribution+confidence+interval&f1=0.99&f=BinomialInterval.c_0.99&f2=100&f=BinomialInterval.n_100&f3=0.03&x=2&y=3&f=BinomialInterval.phat_0.03&a=*FVarOpt.1-_***BinomialInterval.phat--.***BinomialInterval.x---.*--

  762. ⁠, Bret Victor (2011-03-10):

    Do our reading environments encourage active reading? Or do they utterly oppose it? A typical reading tool, such as a book or website, displays the author’s argument, and nothing else. The reader’s line of thought remains internal and invisible, vague and speculative. We form questions, but can’t answer them. We consider alternatives, but can’t explore them. We question assumptions, but can’t verify them. And so, in the end, we blindly trust, or blindly don’t, and we miss the deep understanding that comes from dialogue and exploration.

    Explorable Explanations is my umbrella project for ideas that enable and encourage truly active reading. The goal is to change people’s relationship with text. People currently think of text as information to be consumed. I want text to be used as an environment to think in.

    This essay presents examples of a few initial ideas:

    1. A reactive document allows the reader to play with the author’s assumptions and analyses, and see the consequences….The reader can play with the premise and assumptions of various claims, and see the consequences update immediately. It’s like a spreadsheet without the spreadsheet.
    2. An explorable example makes the abstract concrete, and allows the reader to develop an intuition for how a system works.
    3. Contextual information allows the reader to learn related material just-in-time, and cross-check the author’s claims.
  763. https://www.wsj.com/articles/germany-arrests-suspected-gun-dealer-in-munich-rampage-case-1471372591

  764. http://www.wunderground.com/history/airport/KNHK/2012/7/11/CustomHistory.html?dayend=9&monthend=3&yearend=2013&req_city=NA&req_state=NA&req_statename=NA&MR=1

  765. https://www-cs-faculty.stanford.edu/~knuth/faq.html

  766. https://www.abebooks.com/

  767. https://www.barnesandnoble.com/

  768. https://www.betterworldbooks.com/

  769. http://www.cdc.gov/ncbddd/birthdefects/index.html

  770. http://www.cdc.gov/nchs/births.htm

  771. https://www.cylab.cmu.edu/

  772. https://www.discoverbooks.com

  773. https://www.e-codices.unifr.ch/en/list/one/vad/0296

  774. http://www.genome.gov/pages/der/seqcost2015_4.xlsx

  775. http://www.iarpa.gov/index.php/research-programs/ace

  776. https://www.impactcybertrust.org/dataset_view?idDataset=896

  777. https://michaelnielsen.org/ddi/if-correlation-doesnt-imply-causation-then-what-does/

  778. https://www.moma.org/collection/works/4134

  779. ⁠, Zheng, Yayuan Zhu, Jianhong Zhou, Manru Cui, Liao Yao, Weimin Liu, Yuyu (2013):

    Introduction: It has been suggested that vitamin D is effective to prevent mortality. However, there is no consistent conclusion that the effects of vitamin D supplementation on all-cause mortality are associated with duration of treatment. We conducted a meta-analysis regarding this issue in an effort to provide a more robust answer.

    Methods: A comprehensive search in a number of databases, including MEDLINE, Embase and The Cochrane Central Register of Controlled Trials, was conducted for collecting randomized controlled trials () on vitamin D supplementation preventing mortality. Two investigators independently screened the literature according to the inclusive and exclusive criteria and the relative data were extracted. Data analysis was performed by using Review Manager 5.0 software.

    Results: Data from forty-two RCT s were included. Vitamin D therapy significantly decreased all-cause mortality with a duration of follow-up longer than 3 years with a RR (95% CI) of 0.94 (0.90-0.98). No benefit was seen in a shorter follow-up periods with a RR (95% CI) of 1.04 (0.97-1.12). Results remain robust after sensitivity analysis. The following subgroups of long-term follow-up had significantly fewer deaths: female only, participants with a mean age younger than 80, daily dose of 800 IU or less, participants with vitamin D insufficiency (baseline 25-hydroxyvitamin D level less than 50 nmol/​​​​L) and cholecalciferol therapy. In addition, the combination of vitamin D and calcium significantly reduced mortality and vitamin D alone also had a trend to decrease mortality in a longer time follow up.

    Conclusions: The data suggest that supplementation of vitamin D is effective in preventing overall mortality in a long-term treatment, whereas it is not significantly effective in a treatment duration shorter than 3 years. Future studies are needed to identify the efficacy of vitamin D on specific mortality, such as cancer and cardiovascular disease mortality in a long-term treatment duration.

  780. https://www.tensorflow.org/guide/summaries_and_tensorboard

  781. https://www.theindiaforum.in/article/revolt-upper-castes

  782. ⁠, Thomas Dimson (2020-05-13):

    [ samples generated after training on a dictionary and heavily filtered to try to remove existing words (source). Example:

    pellum (noun)

    the highest or most important point or position

    “he never shied from the pellum or the right to preach”

    Discussion: HN⁠,     /​ ​​ ​​ ​​ ​r    /​ ​​ ​​ ​​ ​ML]

    …Most of the project was spent throwing a number of rejection tricks to make good samples, eg.,

    • Rejecting samples that contain words that are in the a training set / blacklist to force generation completely novel words
    • Rejecting samples without the use of the word in the example usage
    • Running a part of speech tagger on the example usage to ensure they use the word in the correct POS
  783. https://www.thriftbooks.com/

  784. http://www.vocativ.com/interactive/underworld/drugs/darknet-arrests-map/

  785. https://www.writingroutines.com/routines/

  786. https://xkcd.com/chesscoaster/

  787. https://www.yalelawjournal.org/note/amazons-antitrust-paradox

  788. https://news.ycombinator.com/item?id=11646822

  789. ⁠, Brian Hayes (2020-09-04):

    [Series of simulations exploring explanations for the “paradox of the plankton”: why, in apparently homogenous environments, such as open sea water, are there countless thousands of species of plankton all doing the same task of photosynthesis & competing for the same resources? Why isn’t there just a few, or one, species which is optimal in that niche and outcompetes all the others rapidly? Similarly, in a forest: why is there such a dense mix of tree species, rather than relatively continuous stands of species as local conditions vary?

    The initial simulations demonstrate that even a tiny fitness difference will result in a loss of diversion; in fact, with no difference, simple random fluctuations, ‘drift’, will eventually irreversibly drive species to extinction (even assuming occasional ‘immigrants’). This can be fended off by assuming a specialty, like metabolizing a particular chemical well, but can this really explain forest stands with 200+ species? More plausible is predatory-prey dynamics like the Lotka-Volterra model: a species which becomes too common gets preyed on by diseases and parasites and predators, stopping it from spreading further. The dynamics of it are chaotic, but preserve diversity. To some extent, probably all of these explanations are true.]

  790. Backstop

  791. Turing-complete#security-implications

  792. https://pannenkoek2012.fandom.com/wiki/Parallel_Universe

  793. https://www.youtube.com/watch?v=kpk2tdsPh0A

  794. http://beza1e1.tuxen.de/articles/accidentally_turing_complete.html

  795. http://okmij.org/ftp/Computation/sendmail-as-turing-machine.txt

  796. 2006-jared-wikimediaprovesgreenspunstenthlaw.html

  797. http://www.catonmat.net/blog/proof-that-sed-is-turing-complete/

  798. https://github.com/Tegmen/RegEx-BrainF-Interpreter/

  799. https://esolangs.org/wiki////

  800. https://esolangs.org/wiki/Thue

  801. http://citeseerx.ist.psu.edu/viewdoc/download?doi=

  802. http://web.mat.bham.ac.uk/R.W.Kaye/minesw/infmsw.pdf

  803. http://mkv25.net/dfma/map-8269

  804. http://dwarffortresswiki.org/index.php/DF2014:Computing

  805. http://gaming.stackexchange.com/a/20220

  806. http://www.tt-forums.net/viewtopic.php?f=29&t=37902&sid=75000c8f5f3b17f607335077dad6ce94&view=print

  807. http://wiki.openttdcoop.org/Logic

  808. 2019-07-15-danielbali-citiesskylineisturingcomplete.html

  809. ⁠, Radu Grigore (2016-05-17):

    This paper describes a reduction from the halting problem of Turing machines to subtype checking in Java. It follows that subtype checking in Java is undecidable, which answers a question posed by Kennedy and Pierce in 2007. It also follows that Java’s type checker can recognize any recursive language, which improves a result of Gil and Levy from 2016. The latter point is illustrated by a parser generator for fluent interfaces.

  810. ⁠, Hessameddin Akhlaghpour (2020-08-20):

    Life is confronted with computation problems in a variety of domains including animal behavior, single-cell behavior, and embryonic development. Yet we currently have no biologically plausible model capable of universal computation, i.e., Turing-equivalent in scope. Network models (which include neural networks, intracellular signaling cascades, and gene regulatory networks) fall short of universal computation, but are assumed to be capable of explaining cognition and development. I present a class of models that bridge two concepts from distant fields: combinatory logic (or, equivalently, lambda calculus) and molecular biology. A set of basic RNA editing rules can make it possible to compute any computable function with identical algorithmic complexity to that of Turing machines. The models do not assume extraordinarily complex molecular machinery or any processes that radically differ from what we already know to occur in cells. Distinct independent enzymes can mediate each of the rules and RNA molecules solve the problem of parenthesis matching through their secondary structure. The most plausible of these models does not strictly mimic the operation rules of combinatory logic or lambda calculus; it relies on standard RNA transcription from static genomic templates and the editing rules can be implemented with merely cleavage and ligation operations. This demonstrates that universal computation is well within the reach of molecular biology. It is therefore reasonable to assume that life has evolved—or possibly began with—a universal computer that yet remains to be discovered. The variety of seemingly unrelated computational problems across many scales can potentially be solved using the same RNA-based computation system. Experimental validation of this theory may immensely impact our understanding of memory, cognition, development, disease, evolution, and the early stages of life.

  811. http://kristerw.blogspot.com/2016/01/more-turing-completeness-in-surprising.html

  812. http://googleprojectzero.blogspot.com/2015/07/one-font-vulnerability-to-rule-them-all.html

  813. https://scarybeastsecurity.blogspot.com/2016/12/redux-compromising-linux-using-snes.html

  814. https://scarybeastsecurity.blogspot.com/2016/11/0day-exploit-compromising-linux-desktop.html