Hexbyte  Hacker News  Computers Data Mining Reveals the Crucial Factors That Determine When People Make Blunders

Hexbyte Hacker News Computers Data Mining Reveals the Crucial Factors That Determine When People Make Blunders

Hexbyte Hacker News Computers

The way people make decisions in the real world is a topic of increasing interest among psychologists, social scientists, economists, and others. It determines how economies perform, how elections are run, and how conflicts break out and get resolved.

One idea has provided a focal point for decision-making research. This is the notion of bounded rationality—that people are limited by various constraints in the real world, and these play a crucial role in the decision-making process.  People are limited by the difficulty of the decision they have to make, their own decision-making skill, and the time they can spend on the problem. Nevertheless, whatever the circumstances, a decision has to be made and the consequences accepted.

That raises an important set of questions. How do these factors influence the quality of the decision being made? Does time pressure have a bigger impact than, say, decision-making skill on the quality of a decision?

Hexbyte  Hacker News  Computers Data Mining Reveals the Crucial Factors That Determine When People Make Blunders

These are hard questions to answer, given the difficulty of setting up a controlled experiment to test them. Indeed, nobody has found a satisfactory way of studying the problem.

Until now. Today, Ashton Anderson at Microsoft Research in New York City, Jon Kleinberg at Cornell University in Ithaca, and Sendhil Mullainathan at Harvard University in Cambridge unveil the first large-scale study of decision making under controlled conditions. For the first time, these guys have been able to study how the quality of decision making changes with the time available, the skill of the decision maker, and the difficulty of the decision at hand.

Their laboratory? The game of chess. “We have used chess as a model system to investigate the types of features that help in analyzing and predicting error in human decision-making,” they say.

Their research focuses on a database of 200 million chess games played online between amateurs and another database of around one million games played between grand masters. What’s interesting about these databases is that the outcome of the game reveals whether a player has made a mistake. And the recorded moves reveal exactly when the losing player makes the blunder.

The team are then able to see what factors have played a role. They can see whether the player was under time pressure, for example. They can see the difficulty of the decision by examining the position on the board and its complexity. They do this by totting up all possible moves and then working out what fraction of them are blunders. So a position in which all moves except one are blunders is more difficult than a position in which only one move out of many is a blunder.

The team also know the skill level of the players. The skill level of every chess player is given by a number called the Elo rating (after Arpad Elo, who came up with it). Most amateurs have a rating between 1000 and 2000, strong amateurs get up to 2400, and the world’s top players receive rankings of around 2600. There are generally just a handful of players at any time with a ranking over 2800. A difference of 400 points between players suggests that the higher-ranked player is overwhelmingly likely to win.

And the huge size of the database allows them to cut and dice the data in a way that holds two of these variables constant while allowing the other to vary. For example, the team can examine board positions of the same difficulty while players are under the same time pressure to see how any variation in their skill level influences the quality of the decisions they make. Equally, the researchers can hold skill and time pressure constant while allowing board position to vary; and so on.

The results make for interesting reading. They find, for example, that the amount of time spent on a decision is a factor in blundering, but only up to a point. Quick decisions are more likely to lead to a blunder, but after about 10 seconds or so the likelihood of a blunder flattens out. So when players spend more time than this on a move, it is probably because they don’t know what to do.

The difficulty of the decision is an important factor, too. More difficult positions are more likely to lead to a blunder. And skill levels have a big impact in reducing the likelihood of a blunder. In general, better players make better decisions.

But Anderson and co have found evidence of an entirely counterintuitive phenomenon in which skill levels play the opposite role, so that skillful players are more likely to make an error than their lower-ranked counterparts. The team call these “skill anomalous positions.”

That’s an extraordinary discovery which will need some teasing apart in future work. “The existence of skill-anomalous positions is surprising, since there is a no a priori reason to believe that chess as a domain should contain common situations in which stronger players make more errors than weaker players,” say Anderson and co. Just why this happens isn’t clear.

These results have an important application. They allow the team to predict when a player is most likely to make a mistake. And it turns out that one of the factors is a much more powerful predictor than the others.

The bottom line is that the difficulty of the decision is the most important factor in determining whether a player makes a mistake.  In other words, examining the complexity of the board position is a much better predictor of whether a player is likely to blunder than his or her skill level or the amount of time left in the game.

That could have important implications for the way researchers examine other decisions. For example, how does the error rate of highly skilled drivers in difficult conditions compare with that of bad drivers in safe conditions? If the difficulty of the decision is the crucial factor, rather than driver skill, then much more emphasis needs to be placed on this. “We think of inexperienced and distracted drivers as a major source of risk, but how do these effects compare to the presence of dangerous road conditions?” ask Anderson and co.

And given the team’s discovery of skill-anomalous conditions, are there road conditions that make skillful drivers more likely to make a mistake than less skillful ones?

This kind of work will have big implications beyond driving. Economists might well ask what all this means for buying decisions, election officials will ask about the complexity of information related to voting decisions, and negotiators will think about its impact on resolving conflict.

Fascinating work and plenty of food for thought.

Ref: arxiv.org/abs/1606.04956  : Assessing Human Error Against a Benchmark of Perfection

Read More

Hexbyte  Tech News  Wired Tumblr’s Porn Ban Reveals Who Controls What We See Online

Hexbyte Tech News Wired Tumblr’s Porn Ban Reveals Who Controls What We See Online

Hexbyte Tech News Wired

Tumblr was never explicitly a space for porn, but, like most things on the internet, it is chock full of it anyway. Or at least it was. On Monday, to the shock of the millions of users who had used the microblogging site to consume and share porn GIFs, images, and videos, Tumblr banned the “adult content” that its CEO, David Karp, had defended five years prior. In the hours after the announcement, sex workers panicked, users threatened to leave, and—in classic Tumblr fashion—online petitions calling for change gained hundreds of thousands of signatures. But Tumblr’s porn ban isn’t about porn or Tumblr at all, really. It’s about the companies and institutions who wield influence over what does and doesn’t appear online.

When Melissa Drew, an adult content creator and model, logged in to Tumblr Monday afternoon, she was greeted by a deluge of unfamiliar posts and notifications. Her usual feed, perfectly curated after nearly a decade of tinkering, was awash with panicked posts from fellow adult models, memes about the policy change, and goodbye posts. Drew’s personal blog, which she had relied on as a public-facing way to tease the content available to her Patreon subscribers, was lit up with notifications from Tumblr informing her that most of her posts violated the new rules.

When Yahoo bought Tumblr for $1.1 billion in 2013, critics warned that premium advertisers wouldn’t exactly be clamoring to run ads in a sea of porn. Yahoo CEO Marissa Mayer disagreed, arguing that targeting tools would keep content that isn’t “brand-safe” (read: porn) away from ads. In an interview shortly after the acquisition, Karp doubled down on the platform’s openness to porn. “When you have … any number of very talented photographers posting tasteful photography,” said Karp, “I don’t want to have to go in there to draw the line between this photo and the behind the scenes photo of Lady Gaga and like, her nip.”

But the nip line has indeed been drawn, and it’s a doozy. Though Tumblr’s Monday announcement had technically only prohibited depictions of sex acts, “human genitalia,” and “female-presenting nipples,” a much wider swath of Drew’s feed was quickly caught up in the ban. “I had everything from nude, censored nudity, and lingerie photos flagged,” she told WIRED. “I still haven’t dealt with removing all of them yet—I just sort of heavy sighed and closed the tab.”

Safe Space

In interviews and messages with WIRED, more than 30 sex workers, porn consumers, and creators on Tumblr lamented the loss of what they described as a unique, safe space for curated sexually themed GIFs, photos, and videos. Many users who had used the microblogging site as their primary source for porn were at a loss when asked where they would go after Tumblr’s ban on “adult content” goes into effect on December 17. For the thousands of sex workers who used the site to share their own explicit content in a controlled, relatively contained manner—not to mention the countless others who used that content to fill the hyper-curated feeds of some of the site’s most popular porn blogs—the crackdown’s consequences are even more difficult to unpack. And researchers say the ban could shrink Tumblr’s user base, which already appears in turmoil over the decision.

The move comes less than two weeks after Apple pulled Tumblr from the iOS App Store after child pornography was found on the site. Though the offending illegal content was removed quickly, according to Tumblr, the app has yet to return to the App Store (it was never removed from the Google Play Store). In its most recent blog post, Tumblr stated that its longstanding no-tolerance policy against child pornography should not be conflated with the move to ban adult content. The latter, Tumblr argued, was inspired by a drive to create “a better Tumblr.” But these sorts of decisions aren’t made in a vacuum.

In March, Congress passed the Fight Online Sex Trafficking Act and the Stop Online Sex Trafficking Act (or FOSTA-SESTA for short). Lawmakers hailed the law as a means to give prosecutors more tools to combat combat sex trafficking. But the statute also tinkered with a bedrock provision of internet law, opening the door for platforms to be held criminally and civilly liable for the actions of their users. The law’s passage immediately led to the closure of several sex-work-related online venues, such as Craigslist’s personals section, numerous subreddits, and Patreon’s support for adult creators.

In interviews, more than a dozen sex workers told WIRED that the support and openness of adult-content creators on Tumblr had attracted them to the site. They described the site as notably more empowering and friendly than more traditional venues for explicit content, like PornHub, and relished Tumblr’s freedom and opportunity for virality. Many used the site to promote their paid content on other sites, interact with fellow sex workers, and screen clients.

Liara Roux, a sex worker and online political organizer, told WIRED via email that the options for finding adult content online are diminishing, and consolidating with big companies “and away from community generated content and independent creator friendly platforms.” This, Roux says, is dangerous: “As mainstream sites slowly remove sexual content, which often is how queer and other marginalized communities are able to connect, it will become difficult for both sex workers and these communities to have an online space to exist.”

App Store Sway

The ubiquity of the App Store gives Apple considerable influence over online content. Historically, Apple has taken a sex-negative approach to policing, with former CEO Steve Jobs once famously stating that “folks who want porn can buy an Android phone,” but the company’s influence extends beyond the iOS sphere. In the days after it was delisted from the App Store, Tumblr ramped up its moderation efforts, erroneously deleting numerous SFW and NSFW Tumblr accounts unrelated to child exploitation and abuse, and pushed out an update for its Android app that forced all users to have “safe mode” toggled on, restricting access to explicit content. Then came the porn ban.

Apple’s App Store review guidelines prohibit apps “with user-generated content or services that end up being used primarily for pornographic content.” The key word here is “primarily,” as all popular social media platforms are rife with porn, but Apple mostly seems to care about how easily accessible it is. Case in point: Apple pulled a number of Reddit apps from the App Store in 2016, citing issues with an NSFW content toggle, which, when activated, could be used to view pornographic content. The iOS Reddit apps currently available on the App Store have built-in features that prohibit porn subreddits from appearing in search, and make it extremely difficult for users to find NSFW content in-app without labor-intensive workarounds.

Tumblr likely could have instituted a similar content-censorship-based workaround, but it didn’t. Instead, it has gone with what appears to be an all-out-ban of “photos, videos, or GIFs that show real-life human genitals or female-presenting nipples, and any content—including photos, videos, GIFs and illustrations—that depicts sex acts.” Photos of nipples that appear to belong to a female-identifying person—which may be a difficult category to define, judging by the attempts of other platforms, like Instagram, to do the same—may be permitted so long as they are shared as part of a non-sexual context like a post showcasing breastfeeding, pre- or post-birth, post-mastectomy, or gender confirmation surgery. Written erotic content, along with nudity in art—“such as sculptures and illustrations,” says Tumblr—are alright, too. To draw the distinctions, the company says it will use a mix of machine learning and human moderation, and that all appeals for posts erroneously flagged as adult will be reviewed by “real, live human eye(s).”

Tumblr users are already finding problems with the flagging system. Classical paintings of Jesus Christ were flagged, as were photos and GIFs of fully clothed people, patents for footwear, line drawings of landscape scenes, discussions about LGBTQ+ issues and more. “In its early stages, Tumblr is using a poor system for flagging content,” says Abigail Oakley, a researcher at Arizona State University with a focus on Tumblr communities. “Having a certain number of flags on your blog (regardless of their validity) also removes the blog from Google searches, which is another form of censorship.”

Casey Fiesler, an assistant professor at the University of Colorado who specializes in fandom culture on platforms like Tumblr, thinks Tumblr’s crackdown will likely lead to a mass exodus of users. Fielser also worries that the LGBTQ people and sexual assault survivors who use Tumblr as a space for support could inadvertently be affected by the ban. “Even for fandom participants who don’t create adult content themselves, this kind of policy feels like an attack on the community,” she said.

Consumers and creators of porn on Tumblr aren’t entirely sure where they’ll go next. While many sex workers mentioned Reddit and Twitter as two popular alternatives, they lamented the platforms’ lack of community and sex-positive culture. Similarly, more than a dozen people who said they had used Tumblr as their primary source of porn for years praised the site’s social, judgment-free culture, which many cited as helping them understand their sexual orientation. “The technology is out there to allow users to continue browsing a variety of porn from independent producers, but it will require a shift back in how people think of the internet,” said Roux, the sex worker and political organizer. “I’m hoping these kinds of controversies encourage content creators to take as much control over their distribution channels as possible and inspires tech companies to solve the issue of bringing content to users without hosting the content themselves.”


More Great WIRED Stories

Read More

‘Ninjabot’ Reveals the Mantis Shrimp’s Wily Snail-Hunting Scheme

‘Ninjabot’ Reveals the Mantis Shrimp’s Wily Snail-Hunting Scheme

The mantis shrimp is neither a mantis nor a shrimp, but it does wield perhaps the most stunning strike in the animal kingdom. Sitting below its face are two hammers, which the crustacean cocks back and launches at its prey with such speed that it shatters snail shells and tears crabs’ limbs right of their bodies. These things are ornery, and will even fight a human given the chance. For the mantis shrimp, the only tool they have is a hammer, and all the world looks like a nail.

But perhaps that’s not giving the mantis shrimp enough credit. Because like any good scrapper, this critter can’t just throw punches willy nilly—it’s got to strategize. Just how much so, scientists report today in the Journal of Experimental Biology. The first righteous bit: They found that depending on the shape of a marine snail’s shell, a mantis shrimp will change up its punching strategy. And the second bit: The researchers validated these strategies with something called Ninjabot, a machine that replicates the legendary strike of the feistiest crustacean on Earth.

Patek Lab/Duke University

A mantis shrimp’s punch is so fast and powerful, it produces what’s known as cavitation bubbles. When the hammer launches at prey, it develops an area of low pressure, in which vapor bubbles form. These suddenly collapse to release a shockwave that supplements the punch and increases the local temperature up to 8,500 degrees F. That heat is only fleeting, but the strike is so intense, it will often knock prey like crabs out cold.

Snails, though, they’re a whole other challenge for the mantis shrimp. The researchers looked at species with three shell types in particular. The first, known as “low-spired,” is a squat shell with a wide aperture, or the opening where the meat of the snail pokes out. The second, known as “high-spired,” is a much longer, less girthy shell with a small aperture and pointy apex, or tip. And the third is something in between, with a wide aperture like the first, only a bit lengthier.

When presented with a snail with a wider opening, a mantis shrimp attacks that bit relentlessly. “Overwhelmingly they’re just hitting the aperture the whole time,” says Stanford University biologist Rachel Crane, lead author on the new paper. “They rarely hit the apex, and then they also avoid the whorl.” The whorl being the center bit, basically everything that’s not the opening or the tip.

“The story with the higher-spired shells is more complicated,” says Crane. In general, the mantis shrimp tends to start by hitting the opening like it did with the squatter, low-spired shells. But as they work at the shell, they switch to striking that pointy tip. “Which is weird,” Crane adds. “Based on how cracks spread through a shell, it’s unlikely that strikes at the aperture would weaken the apex.”

So what’s going on here? Enter Ninjabot. The machine is set up much like an actual mantis shrimp appendage. It’s got a spring, which stores energy, a hammer, which does the hammering, obviously, and a latch, which lets loose the hammer. With this machine, the researchers could replicate the same forces of a mantis shrimp punch on different parts of the shells, only in a more controlled way. (Good luck asking a mantis shrimp to punch on command.)

What they found is that—as the mantis shrimp’s behavior suggests—the aperture is the sweet spot. “When Ninjabot hit the aperture the first couple of times, the shell would usually break off a little bit, so pieces of the lip would chip away,” says Crane.

But. That success rate plummeted pretty rapidly on high-spired shells with relatively small apertures. “It seems that the aperture with these high-spired shells is a strategy of diminishing returns,” Crane adds. “So they start off and it seems like it’s going to be really successful, but likely due to aspects of the shell’s morphology, that gets less and less successful over time.” So the predator switches from striking the opening to striking the tip.

What the mantis shrimp definitely doesn’t do, though, is switch and strike at the whorl. It’s pretty much indestructible for our boxer, which the researchers also confirmed with Ninjabot. “The easy answer is that the whorls are terrible,” says Crane. “The mantis shrimp never hit them.”

So the mantis shrimp doesn’t just come across a snail and bash the hell out of it at random. It customizes its strategy based on the shape of the shell, like a boxer adapting to different opponents. And like a good boxer, it’s extremely perceptive about its target. Crane isn’t sure yet how exactly the mantis shrimp sizes up the shell—it could be sight or feel or even chemosensory cues. But what’s clear is the predator deliberately takes the snail into its maxillipeds, small appendages it uses to feed, rolling the shell around before jamming it into the substrate and making its strike, which lasts mere milliseconds.

Really, that makes the mantis shrimp more like a patient archer than a boxer. All the aiming and planning and setting up has to happen before the critter releases its strike—it can’t adjust once the hammers are let loose. “It takes a really long time in order for an animal to be this fast,” Crane says. It’s persistent too: The researchers caught one mantis shrimp hitting a shell 460 times.

Far from being a “brute” that its power would suggest, the creature relies on complex behaviors that it adapts to different prey. All the world may still be a nail to the mantis shrimp, but it’s no mere hammer.

Read More