Groups Similar Look up By Text Browse About

Similar articles
Article Id Title Prob Score Similar Compare
143037 VENTUREBEAT 2019-7-11:
Facebook’s AI beats human poker champions
1.000 Find similar Compare side-by-side
143039 THEVERGE 2019-7-11:
Facebook and CMU’s ‘superhuman’ poker AI beats human pros
0.990 0.683 Find similar Compare side-by-side
142999 ARSTECHNICA 2019-7-11:
Facebook AI Pluribus defeats top poker professionals in 6-player Texas Hold ’em
0.985 0.653 Find similar Compare side-by-side
143004 ENGADGET 2019-7-11:
Facebook and CMU's poker AI beat five pros at once
0.985 0.592 Find similar Compare side-by-side
142924 TECHCRUNCH 2019-7-11:
AI smokes 5 poker champs at a time in no-limit Hold’em with ‘relentless consistency’
0.975 0.565 Find similar Compare side-by-side
142922 VENTUREBEAT 2019-7-11:
Facebook VP: AI has a compute dependency problem
0.324 0.516 Find similar Compare side-by-side
142974 ENGADGET 2019-7-11:
DeepMind's ‘Starcraft II’ AI will play public matches
0.017 0.508 Find similar Compare side-by-side
143226 VENTUREBEAT 2019-7-12:
AI Weekly: Highlights from VentureBeat’s AI conference Transform
0.006 0.501 Find similar Compare side-by-side
142995 ARSTECHNICA 2019-7-11:
DeepMind AI is secretly lurking on the public StarCraft II 1v1 ladder
0.014 0.479 Find similar Compare side-by-side
142436 VENTUREBEAT 2019-7-8:
The AI for Good nominees for the Transform AI Innovation Awards
0.478 Find similar Compare side-by-side
142914 VENTUREBEAT 2019-7-11:
Announcing the AI Innovation Awards winners at Transform 2019
0.017 0.458 Find similar Compare side-by-side
143369 TECHREPUBLIC 2019-7-15:
Why companies plan to double AI projects in the next year
0.451 Find similar Compare side-by-side
143123 VENTUREBEAT 2019-7-12:
To be successful with AI, you have to start small
0.012 0.444 Find similar Compare side-by-side
142767 VENTUREBEAT 2019-7-10:
OpenAI: Explainability and reasoning should inform future AI models
0.006 0.437 Find similar Compare side-by-side
143102 VENTUREBEAT 2019-7-11:
AWS AI VP: Developers drive all innovation in technology
0.003 0.421 Find similar Compare side-by-side
142524 VENTUREBEAT 2019-7-8:
IDC: For 1 in 4 companies, half of all AI projects fail
0.417 Find similar Compare side-by-side
142559 VENTUREBEAT 2019-7-9:
The Startup Spotlight nominees for the Transform AI Innovation Awards
0.004 0.413 Find similar Compare side-by-side
143104 TECHCRUNCH 2019-7-11:
Andrew Ng to talk about how AI will transform business at TC Sessions: Enterprise
0.003 0.405 Find similar Compare side-by-side
142880 VENTUREBEAT 2019-7-10:
Airbnb Homes CTO: There’s AI talent at community colleges, not just elite colleges
0.002 0.392 Find similar Compare side-by-side
142938 VENTUREBEAT 2019-7-11:
Intel: New chip materials will enable massive AI research gains
0.006 0.391 Find similar Compare side-by-side
142791 TECHREPUBLIC 2019-7-10:
70% of US employees no longer fear AI
0.388 Find similar Compare side-by-side
143351 VENTUREBEAT 2019-7-15:
How video game engines help create smarter AI
0.385 Find similar Compare side-by-side
143157 VENTUREBEAT 2019-7-12:
Winners of VentureBeat’s first Women in AI Awards announced at Transform 2019
0.371 Find similar Compare side-by-side
142894 VENTUREBEAT 2019-7-10:
Takeaways from the inaugural Women in AI Breakfast at Transform 2019
0.371 Find similar Compare side-by-side
142740 VENTUREBEAT 2019-7-10:
Google Cloud’s Andrew Moore says AI must be more accessible to succeed
0.005 0.364 Find similar Compare side-by-side


ID: 143037


Date: 2019-07-11

Facebook’s AI beats human poker champions

Facebook AI Research and Carnegie Mellon University today detailed the creation of Pluribus, a poker-playing AI that Facebook claims is the first to beat professionals in Texas Holdem. Poker is a game sometimes used to benchmark the performance of artificial intelligence or game theory. The bot bested 15 human professionals, all of whom have won at least $1 million playing poker, such as 2000 World Series of Poker champion Chris Jesus Ferguson and 4-time World Poker Tour champion Darren Elias. Facebook AI claims Pluribus is the first AI to consistently beat more than 2 human players in a benchmark game. How the AI was constructed was detailed in a paper published in the journal Science. Researchers say its able to surpass top human performance within 20 hours of training. Pluribus achieves its goal with just $150 worth of cloud computing for training. Like AI trained to play games like Go, Dota II and StarCraft II that precede Pluribus, the AI achieves its results by training in matches against itself. Training for more than 20 hours produces an AI player better than top human players, researchers said. The core of Pluribuss strategy was computed via self play, in which the AI plays against copies of itself, without any data of human or prior AI play used as input. The AI starts from scratch by playing randomly, and gradually improves as it determines which actions, and which probability distribution over those actions, lead to better outcomes against earlier versions of its strategy, the Science paper reads. The AI is Is a collaboration between Carnegie Mellon University Computer Science Department and Facebook AI Research, along with companies like Strategic Machine, Strategy Robot, and Optimized Markets. In tests by researchers, Pluribus won in both 5 humans and 1 AI matches as well as 5 AI and 1 human games. If each chip was worth $1, Pluribus would have made approximately $5 on each hand and earned roughly $1,000/hour playing against 5 humans, Facebook AI said. The exact number of bets it considers varies between one and 14 depending on the situation. Although Pluribus can limit itself to only betting one of a few different sizes between $100 and $10,000, when actually playing no-limit poker, the opponents are not constrained to those few options, the Science paper reads. Pluribus builds on Libratus, an AI poker player made by Carnegie Mellon in 2017, but it comes with some additional features, like a search algorithm to evaluate outcomes a few moves ahead. Abstraction is also used to reason about betting in future rounds and to batch strategically similar hands. The AI also uses counterfactual regret  minimization, an iterative self-play algorithm that plays against itself for improvements.