Libratus is an artificial intelligence computer program designed to play poker, specifically heads up no-limit Texas hold 'em. If you wondered about heads up LHE, forget it. It's solved by 100 per cent. But there is still a long way to go for NLHE and once again Carnegie Mellon challenges the best heads up pros to take on their artificial intelligence.
- Game no limit texas holdem HU blinds 50/100
- 20000 starting stack (each hand starts with 20000 in chips no matter what)
Jason Les, Dong Kyu Kim, Daniel McAulay and Jimmy Chou — kept things reasonably close at the start but their gains were wiped out by an ill-fated change in strategy and forced them to chase the AI for the remaining weeks. At the end of day 20 and after 120,000 players, Libratus claimed victory in theoretical chips with a daily average of $206,061, and an overall pile of $1,766,250. Which makes Libratus 14.71875 bb/100 winner.
See more: How solvers revolutionise poker strategy
FAQ
Q: What is the Brains vs. AI Challenge?
A: The Brains vs. AI Challenge was a competition in 2017 where four professional poker players competed against an AI program called Libratus, developed by Carnegie Mellon University.
Q: What was the competition format?
A: The competition was a variant of Texas Hold'em, where each player had 30,000 chips and played a total of 120,000 hands over 20 days. The competition was designed to test the AI's ability to adapt to different playing styles and make strategic decisions in complex situations.
Q: Who participated in the competition?
A: The competition featured four of the world's best professional poker players: Dong Kim, Jason Les, Jimmy Chou, and Daniel McAulay.
Q: Who won the competition?
A: The AI program Libratus won the competition, finishing with a lead of $1.7 million in chips over the four human players.
Q: How did the AI program perform so well?
A: The AI program used a combination of techniques, including game theory and machine learning, to analyze the players' strategies and adapt its own strategy accordingly. It also employed a "nested subgame solving" approach that allowed it to make decisions based on multiple layers of strategy.
Q: What were the implications of the competition?
A: The Brains vs. AI Challenge demonstrated the significant progress that AI has made in the field of game-playing, and the potential for AI to be applied to other complex decision-making domains. It also highlighted the need for continued development of ethical guidelines for the use of AI in decision-making contexts.
Q: Are there plans for future Brains vs. AI Challenges?
A: There have been no announcements of future Brains vs. AI Challenges, but given the success of the 2017 competition, it's possible that similar events could be held in the future.
Comments: