{"id":13541,"date":"2017-02-01T00:08:15","date_gmt":"2017-02-01T00:08:15","guid":{"rendered":"http:\/\/www.kurzweilai.net\/?p=292227"},"modified":"2017-02-02T09:25:52","modified_gmt":"2017-02-02T09:25:52","slug":"carnegie-mellon-ai-beats-top-poker-pros-a-first","status":"publish","type":"post","link":"https:\/\/hoo.central12.com\/fugic\/2017\/02\/01\/carnegie-mellon-ai-beats-top-poker-pros-a-first\/","title":{"rendered":"Carnegie Mellon AI beats top poker pros — a first"},"content":{"rendered":"
\"\"

“Brains vs Artificial Intelligence” competition at the Rivers Casino in Pittsburgh (credit: Carnegie Mellon University)<\/p><\/div>\n

Libratus, an AI developed by Carnegie Mellon University<\/a>, has defeated four of the world\u2019s best professional poker players in a marathon 120,000 hands of Heads-up, No-Limit Texas Hold\u2019em<\/a> poker played over 20 days, CMU announced\u00a0today (Jan. 31) — joining Deep Blue (for chess), Watson, and Alpha Go as major milestones in AI.<\/p>\n

Libratus led the pros by a collective $1,766,250 in chips.* The tournament\u00a0was held at the Rivers Casino in Pittsburgh from 11–30 January in a competition called \u201cBrains Vs. Artificial Intelligence: Upping the Ante.\u201d<\/p>\n

The developers of Libratus — Tuomas Sandholm, professor of computer science, and Noam Brown, a Ph.D. student in computer science — said the sizable victory is statistically significant and not simply a matter of luck. \u201cThe best AI\u2019s ability to do strategic reasoning with imperfect information has now surpassed that of the best humans,\u201d Sandholm said. “This is the last frontier, at least in the foreseeable horizon, in game-solving in AI.”<\/p>\n

This new AI milestone has implications for any realm in which information is incomplete and opponents sow misinformation, said Frank Pfenning, head of the Computer Science Department in CMU\u2019s School of Computer Science. Business negotiation, military strategy, cybersecurity, and medical treatment planning could all benefit from automated decision-making using a Libratus-like AI.<\/p>\n

\u201cThe computer can\u2019t win at poker if it can\u2019t bluff,\u201d Pfenning explained. \u201cDeveloping an AI that can do that successfully is a tremendous step forward scientifically and has numerous applications. Imagine that your smartphone will someday be able to negotiate the best price on a new car for you. That\u2019s just the beginning.\u201d<\/p>\n

How the pros taught Libratus about its weaknesses<\/strong><\/p>\n

\"\"

Brains vs AI scorecard (credit: Carnegie Mellon University)<\/p><\/div>\n

So how was Libratus was able to improve day to day during the competition? It turns out it was the pros themselves who taught Libratus about its weaknesses. \u201cAfter play ended each day, a meta-algorithm analyzed what holes the pros had identified and exploited in Libratus\u2019 strategy,\u201d Sandholm explained. \u201cIt then prioritized the holes and algorithmically patched the top three using the supercomputer each night.<\/p>\n

“This is very different than how learning has been used in the past in poker. Typically researchers develop algorithms that try to exploit the opponent\u2019s weaknesses. In contrast, here the daily improvement is about algorithmically fixing holes in our own strategy.\u201d<\/p>\n

Sandholm also said that Libratus\u2019 end-game strategy was a major advance. \u201cThe end-game solver has a perfect analysis of the cards,\u201d he said. It was able to update its strategy for each hand in a way that ensured any late changes would only improve the strategy. Over the course of the competition, the pros responded by making more aggressive moves early in the hand, no doubt to avoid playing in the deep waters of the endgame where the AI had an advantage, he added.<\/p>\n

Converging high-performance computing and AI<\/strong><\/p>\n

\"\"

Professor Tuomas Sandholm, Carnegie Mellon School of Computer Science, with the Pittsburgh Supercomputing Center’s Bridges supercomputer (credit: Carnegie Mellon University)<\/p><\/div>\n

Libratus\u2019 victory was made possible by the Pittsburgh Supercomputing Center\u2019s Bridges computer<\/a>. Libratus recruited the raw power of approximately 600 of Bridges\u2019 846 compute nodes. Bridges’ total speed is 1.35 petaflops, about 7,250 times as fast as a high-end laptop, and its memory is 274 terabytes, about 17,500 as much as you\u2019d get in that laptop. This computing power gave Libratus the ability to play four of the best Texas Hold\u2019em players in the world at once and beat them.<\/p>\n

\u201cWe designed Bridges to converge high-performance computing and artificial intelligence,\u201d said Nick Nystrom, PSC\u2019s senior director of research and principal investigator for the National Science Foundation-funded Bridges system. \u201cLibratus\u2019 win is an important milestone toward developing AIs to address complex, real-world problems. At the same time, Bridges is powering new discoveries in the physical sciences, biology, social science, business and even the humanities.\u201d<\/p>\n

Sandholm said he will continue his research push on the core technologies involved in solving imperfect-information games and in applying these technologies to real-world problems. That includes his work with Optimized Markets, a company he founded to automate negotiations.<\/p>\n

\u201cCMU played a pivotal role in developing both computer chess, which eventually beat the human world champion, and Watson, the AI that beat top human Jeopardy! competitors,\u201d Pfenning said. \u201cIt has been very exciting to watch the progress of poker-playing programs that have finally surpassed the best human players. Each one of these accomplishments represents a major milestone in our understanding of intelligence.<\/p>\n

Head\u2019s-Up No-Limit Texas Hold\u2019em is a complex game, with 10160<\/sup> (the number 1 followed by 160 zeroes) information sets — each set being characterized by the path of play in the hand as perceived by the player whose turn it is. The AI must make decisions without knowing all of the cards in play, while trying to sniff out bluffing by its opponent. As \u201cno-limit\u201d suggests, players may bet or raise any amount up to all of their chips.<\/p>\n

Sandholm will be sharing Libratus\u2019 secrets now that the competition is over, beginning with invited talks at the Association for the Advancement of Artificial Intelligence meeting<\/a> Feb. 4–9 in San Francisco and in submissions to peer-reviewed scientific conferences and journals.<\/p>\n

* The pros — Dong Kim, Jimmy Chou, Daniel McAulay and Jason Les — will split a $200,000 prize purse based on their respective performances during the event. McAulay, of Scotland, said Libratus was a tougher opponent than he expected, but it was exciting to play against it. \u201cWhenever you play a top player at poker, you learn from it,\u201d he said.<\/em><\/p>\n