Tic-tac-toe fell in 1952, checkers in 1994, chess in 1997 and it now looks like Go, the ancient Chinese game that has a search space many, many times greater than chess, has fallen to a new AI from Google.
…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.
Importantly, AlphaGo isn’t based primarily on searching a huge space but on deep neural networks that learned first from human players and then from simulated play with itself. The techniques, therefore, are not limited to Go.
AlphaGo will face its greatest challenge in March.
AlphaGo’s next challenge will be to play the top Go player in the world over the last decade, Lee Sedol. The match will take place this March in Seoul, South Korea.
Win or lose, I will bet that Lee Sedol is the last human champion the world will ever know.
Max Mendez Beck emails me:
Given the advent of statistics in sports that occurred in the last five years, I am struck by how well soccer works as a metaphor for current epistemological debates regarding the use (and primacy) of quantitative versus qualitative data in social science research. While the three major American sports (football, basketball, and baseball) have been overtaken by a quantitative obsession (count how many tables and numbers you see on an average ESPN show), soccer is emblematic of a sport that is quite difficult to measure quantitatively.
Consider how easy it is to determine who did well in an average NBA game without needing to even watch it. You can just look at points, assists, rebounds, steals, turnovers, etc. In soccer, individual statistics are almost nonexistent. Even as major sports channels have attempted to incorporate quantitative measures into their soccer broadcasts–for example, by showing the number of kilometers a player has covered when he gets subbed out (a pretty uninformative statistic on its own)–these numbers have not caught on with the regular fan.
While in basketball everyone debates about who “the best ever” is by referring to their career averages in points, field goal percentage, PER, etc. In soccer the only statistic that is ever used is goals scores, and goals scored is only one small dimension of a player, even smaller if he is not a striker. It would be silly to judge Andrés Iniesta or Zinedine Zidane on how many goals they scored in a season.
So what is it about soccer that makes it so hard to quantify? Or what makes American sports so easy to measure? One obvious answer is the length of the units that can be easily separated and analyzed. In basketball its a maximum of 24 seconds, in baseball its essentially a pitch (or an at bat), and in football its each snap. For soccer, the only apparent unit to separate out is the 45 minute halftime mark. Changes in possession could be another measure, but even then a team’s single possession could be several minutes long.
However, the real challenge comes in measuring individual accomplishments. Just recently I was watching a Barcelona game and Iniesta clearly was having an amazing game (as was mentioned several times by the announcer), and yet the things that made him have a great game were only describable in words and not numbers. There was a beautiful and sudden “regate” or dribble around a defender before he passed it on to a teammate for a quick counter attack. There was the beautiful pass between defenders that led to an assist for the first goal. There was the sudden change in direction and over the top pass to the other side of the field that put the defenders on their heels. Many of these moves are incredibly situational; they have to do with the rhythm of the game and the need to speed it up or slow it down. Nothing in the boxscore could truly capture these attributes.
So the question is: Is soccer something that can’t be measured in numbers?
Here are various readings on the topic.
I love this scene from Digital Gold, Nathaniel Popper’s entertaining book on the history of and people behind Bitcoin. Wences Casares, a successful internet entrepreneur and Bitcoin enthusiast, is at a party of millionaires and billionaires and he wants to impress the crowd:
To prove how easy this all was, Wences asked Blodget to take out his phone and helped him to create an empty Bitcoin wallet. Once it was up, and Wences had Blodget’s new Bitcoin address, Wences used the wallet on his own phone to send Blodget $250,000..the money was then passed to the phones of other people around the table once they had set up wallets. Anyone could have run off with Wences’s $250,000, but that wasn’t a risk with this particular crowd. Instead, as the money went around, Wences saw the guests’ laughter and wide-eyed amazement at what they were watching.
Wences is something of a character. Russ Roberts did a good interview with him on EconTalk.
Under one model, Tinder teaches you the joys of tussling with those from “the other side of the tracks,” and pulls you away from marrying a fellow Ph.d. — “once you’ve tried Mack, you’ll never go back.”
My intuition differs. You use Tinder in bars, venues, and neighborhoods you have chosen. So you end up tussling with, or mating with, or just chatting with, the more attractive members of your own preferred socioeconomic group. If your group wasn’t on average so sexy to begin with, well at least at the top end it just got a big upgrade in terms of your actual access to attractiveness. So on net the high socioeconomic groups become sexier, at least for those “at the top” with the most choice.
(I am assuming by the way that male photos can to some extent signal status, income, and education, and not just looks; furthermore the male follow-up can demonstrate this readily. And many “connection” services post this information in one form or another as part of the profile. If need be, in general equilibrium bars can adjust their exclusiveness levels to match better to a world-with-Tinder, so that bar patrons are not lured into socioeconomically mistaken “honey pot” marriages.)
Most of all, Tinder gets you out more. You sample more people, even if you don’t end up meeting them through the Tinder app itself. Going to a bar or public space is a better way to spend time than before, and that draws others out too. That’s right, “thick market externalities.” The resulting extra meetings tend to favor assortative mating, just compare such plenitude to a corner solution where you meet only one potential spouse your entire life, namely the proverbial girl next door.
Put it this way: George Clooney or a Silicon Valley billionaire can do better — especially better, compared to others — choosing from 500 people than from five. He (she) has a very good chance of getting his (her) absolute top favorite pick, or close to it. The local milkman also does better from a larger sample size, if only because of match and compatibility issues, but can’t expect to move up so much and of course the pool as a whole can’t “move up” at all. (If you wish, break this down into a positive-sum compatibility component and a competitive zero-sum component; unlike Clooney the milkman may not gain on the latter.)
Finally, Tinder may make it easier for married people to find casual sex, again if they have the right qualifications. Therefore those marrieds may, earlier on, decide to choose a spouse on the grounds of IQ and education, again boosting assortative mating in terms of those features.
In sum, I expect Tinder to boost assortative mating, at least at the top end of the distribution in terms of IQ and education.
And please note, I suspect this increase in assortative mating is a good thing. The abilities of top achievers have a disproportionate impact on the quality of our lives, due to innovation being a public good.
In any case, file under speculative.
Addendum: An interesting twist on the model is to assume that men have some willingness to marry down in terms of education, in return for beauty or other forms of household production, but women do not. An increase in the total sampling of potential partners therefore boosts the marriage prospects of very beautiful women, at the expense of less beautiful women of a given level of educational attainment. In percentage terms, very beautiful women of decent but not extraordinary educational achievement gain the most. Men who are indifferent to such forms of female beauty end up with the smartest children.
I can’t say I understand this FT article so well, but I suppose that is the point. Which are two groups/persons implicated in buying oil from ISIS, or otherwise enabling such trades to take place?
First, Syria. Or is that “Syria.”
Second, the head of the world chess federation, namely Kirsan Ilyumzhinov: “he is best known for his belief in aliens — he has repeatedly recounted an instance when he was abducted in 1997 by “people in yellow spacesuits”.” And this:
Mr Ilyumzhinov has a diverse business empire, stretching from sugar to banking, and a network of contacts to match. He regularly meets the Dalai Lama, and he played chess with Libyan president Muammer Gaddafi shortly before his overthrow.
The paper is by David Hugh-Jones, and this is from the research summary:
The study examined whether people from different countries were more or less honest and how this related to a country’s economic development. More than 1500 participants from 15 countries took part in an online survey involving two incentivised experiments, designed to measure honest behaviour.
Firstly, they were asked to flip a coin and state whether it landed on ‘heads’ or ‘tails’. They knew if they reported that it landed on heads, they would be rewarded with $3 or $5. If the proportion reporting heads was more than 50 per cent in a given country, this indicated that people were being dishonest…
The countries studied – Brazil, China, Greece, Japan, Russia, Switzerland, Turkey, the United States, Argentina, Denmark, the United Kingdom, India, Portugal, South Africa, and South Korea – were chosen to provide a mix of regions, levels of development and levels of social trust.
The study’s author Dr David Hugh-Jones, of UEA’s School of Economics, found evidence for dishonesty in all the countries, but that levels varied significantly across them. For example, estimated dishonesty in the coin flip ranged from 3.4 per cent in the UK to 70 per cent in China. In the quiz, respondents in Japan were the most honest, followed by the UK, while those in Turkey were the least truthful. Participants were also asked to predict the average honesty of those from other countries by guessing how many respondents out of 100 from a particular country would report heads in the coin flip test. However, participants’ beliefs about other countries’ honesty did not reflect reality.
This is interesting:
“Differences in honesty were found between countries, but this did not necessarily correspond to what people expected,” he said. “Beliefs about honesty seem to be driven by psychological features, such as self-projection. Surprisingly, people were more pessimistic about the honesty of people in their own country than of people in other countries.
And consider this from Hugh Jones:
“I suggest that the relationship between honesty and economic growth has been weaker over the past 60 years and there is little evidence for a link between current growth and honesty,” said Dr Hugh-Jones. “One explanation is that when institutions and technology are underdeveloped, honesty is important as a substitute for formal contract enforcement. Countries that develop cultures putting a high value on honesty are able to reap economic gains. Later, this economic growth itself improves institutions and technology, making contracts easier to monitor and enforce, so that a culture of honesty is no longer necessary for further growth.”
The research paper is here, and for the pointers I thank Charles Klingman and Samir Varma.
Jones, by the way, makes it clear there are a variety of kinds of honesty, and inferences from any single test should be limited. For Japan in particular the measured level of honesty depends critically on which test is applied. The real lesson of the study may simply be that most groups are dishonest, and people are not even honest with themselves about their views of the dishonesty of others. Honesty depends a good deal on context too. On some of these questions, see some skeptical comments from Kevin Drum.
If you are looking for simple correlations: “…at individual level, there is no evidence that religious adherence is associated with honesty.” How about having a Ph.d.?
A man who sold himself a $1,000,000 winning D.C. Lottery ticket is just one of many retailers a WUSA9 investigation found winning the lottery at rates statisticians say border on impossible.
At least three retailers won the lottery around 100 times according to an analysis of D.C. Lottery records obtained under the Freedom of Information Act.
“$10,000, $5,000,” Lounes Issaad said about some of his 27 payouts that averaged $30,000 each. “I don’t have nothing to hide.”
…Our investigation found lottery retailers make up at least three of the top five D.C. Lottery frequent winners – all with about 100 wins or more.
You can already rate restaurants, hotels, movies, college classes, government agencies and bowel movements online.
So the most surprising thing about Peeple — basically Yelp, but for humans — may be the fact that no one has yet had the gall to launch something like it.
When the app does launch, probably in late November, you will be able to assign reviews and one- to five-star ratings to everyone you know: your exes, your co-workers, the old guy who lives next door. You can’t opt out — once someone puts your name in the Peeple system, it’s there unless you violate the site’s terms of service. And you can’t delete bad, inaccurate or biased reviews — that would defeat the whole purpose.
Imagine every interaction you’ve ever had suddenly open to the scrutiny of the Internet public.
Is there a word in the English language that more reliably means its opposite than ‘amicable’?
Twitter responses included: “moot,” “humbled,” “nice,” “my friend,” “nonplussed,” “cordial,” “priceless,” “tolerance,” “literally,” “spry,” “sincerely,” “honest,” “pal,” “sure,” and “”Fine” particularly when given as a one word answer.”
My favorite was “spry.” Is there a word for such words? Are there other examples?
Observers seem to focus on the target event and not its complement. Bagchi and Ince have a new paper on this question:
Consumers routinely rely on forecasters to make predictions about uncertain events (e.g., sporting contests, stock fluctuations). The authors demonstrate that when forecasts are higher versus lower (e.g., a 70% vs. 30% chance of team A winning a game) consumers infer that the forecaster is more confident in her prediction, has conducted more in-depth analyses, and is more trustworthy. The prediction is also judged as more accurate. This occurs because forecasts are evaluated based on how well they predict the target event occurring (team A winning). Higher forecasts indicate greater likelihood of the target event occurring, and signal a confident analyst, while lower forecasts indicate lower likelihood and lower confidence in the target event occurring. But because, with lower forecasts, consumers still focus on the target event (and not its complement), lower confidence in the target event occurring is erroneously interpreted as the forecaster being less confident in her overall prediction (instead of more confident in the complementary event occurring—team A losing). The authors identify boundary conditions, generalize to other prediction formats, and demonstrate consequences.
Of course this also has relevance for the evolutionary processes governing pundits.
Here is a related press release (pdf). For the pointer I thank Charles Klingman.
This Connie Chan article about mobile in China is difficult to summarize or excerpt, but it is one of the most important pieces of the year so far. Here are the opening bits:
This post is all about WeChat, but it’s also about more than just WeChat. While seemingly just a messaging app, WeChat is actually more of a portal, a platform, and even a mobile operating system depending on how you look at it.
Much has been written about WeChat in the context of messaging app trends, but few outside of China really understand how it works — and how it can pull off what for many companies (and countries) is still a far-off vision of a world managed entirely through our smartphones. Many of WeChat’s most interesting features — such as access to city services — are not even visible to users outside of China.
So why should people outside of China even care about WeChat? The first and most obvious reason is that it points to where Facebook and other messaging apps could head. Second, WeChat indicates where the future of mobile commerce may lie. Third, WeChat shows what it’s like to be both a platform and a mobile portal (what Yahoo could have been).
Ultimately, however, WeChat should matter to all of us because it shows what’s possible when an entire country — which currently has a smartphone penetration of 62% (that’s almost 1/3 of its population) — “leapfrogs” over the PC era directly to mobile. WeChat was not a product that started as a website and then was adapted for mobile, it was (to paraphrase a certain movie) born into it, molded by it.
Do read the whole thing, this is also one of China’s first major innovations in the classic sense of that term.
The subtitle for this one suffices for a review:
A Soccernomics Guide: Why Chievo Verona, Unterhaching, and Scunthorpe United Will Never Win the Champions League, Why Manchester City, Roma, and Paris Saint-Germain can, and why Real Madrid, Bayern Munich, and Manchester United Cannot Be Stopped
I haven’t even read the full subtitle yet (I used Control C), much less the book. But the author is the highly regarded Stefan Szymanski, and John Foot gave it a very positive review in the 24 July 2015 TLS.
Here is a Schneier on Security post in toto, I won’t indent it once again:
Expertise literature in mainstream cognitive psychology is rarely applied to criminal behaviour. Yet, if closely scrutinised, examples of the characteristics of expertise can be identified in many studies examining the cognitive processes of offenders, especially regarding residential burglary. We evaluated two new methodologies that might improve our understanding of cognitive processing in offenders through empirically observing offending behaviour and decision-making in a free-responding environment. We tested hypotheses regarding expertise in burglars in a small, exploratory study observing the behaviour of ‘expert’ offenders (ex-burglars) and novices (students) in a real and in a simulated environment. Both samples undertook a mock burglary in a real house and in a simulated house on a computer. Both environments elicited notably different behaviours between the experts and the novices with experts demonstrating superior skill. This was seen in: more time spent in high value areas; fewer and more valuable items stolen; and more systematic routes taken around the environments. The findings are encouraging and provide support for the development of these observational methods to examine offender cognitive processing and behaviour.
The lead researcher calls this “dysfunctional expertise,” but I disagree. It’s expertise.
Claire Nee, a researcher at the University of Portsmouth in the U.K., has been studying burglary and other crime for over 20 years. Nee says that the low clearance rate means that burglars often remain active, and some will even gain expertise in the crime. As with any job, practice results in skills. “By interviewing burglars over a number of years we’ve discovered that their thought processes become like experts in any field, that is they learn to automatically pick up cues in the environment that signify a successful burglary without even being aware of it. We call it ‘dysfunctional expertise,'” explains Nee.
See also this paper.”
The pointer is from the estimable Chug.
The newspaper header is:
Panos Kammenos, Greece’s defence minister, threatens to open country’s borders to refugees – including potential members of Islamic State of Iraq and the Levant (Isil) – unless Athens receives debt crisis support
The story is here, via Andrea Castillo. Whether it jives with your mood affiliation or not, it’s time to admit “these people simply don’t know what they are doing.” Fortunately, it does seem the Greek government has been walking back on this talk.
It’s in an economics journal, so it must be true. Agne Suziedelyte reports, from Economic Inquiry:
According to the literature, video game playing can improve such cognitive skills as problem solving, abstract reasoning, and spatial logic. I test this hypothesis using the data from the Child Development Supplement to the Panel Study of Income Dynamics. The endogeneity of video game playing is addressed by using panel data methods and controlling for an extensive list of child and family characteristics. To address the measurement error in video game playing, I instrument children’s weekday time use with their weekend time use. After taking into account the endogeneity and measurement error, video game playing is found to positively affect children’s problem solving ability. The effect of video game playing on problem solving ability is comparable to the effect of educational activities. (JEL I2, J13, J24)
I wonder how much endogeneity can be overcome in such settings, but if nothing else there is positive selection into video games and perhaps you should not be upset if your child is playing them.
Do any of you see an ungated copy? The pointer here is from Kevin Lewis.