A breif summay about Superhuman AI for heads-up no-limit poker: Libratus beats top professionals on Science.
A paper published on Science 2017-12-17 says Libratus(AI) beats top professionals in poker.
Summary
a. AI has defeated top human players in checkers, chess, Go which are all perfect-information games, you can konw all states on the board.
b. Now Libratus(an AI) has defeated top humman players in Texas hold’em(德州扑克). Poker game is an imperfect-information game in which some information about the state of the game is hidden, for example you can not see your opponent's cards holding in his hand.
c. Making strategy under imperfect-information can be used in many fields such as business strategy, negotiation, strategic pricing, finance, cybersecurity, and military applications and so on.
d. Techniques in Libratus do not use expert domain knowledge or human data and are not specific to poker; thus they apply to a host of imperfect-infromation games. Furthur the techniques can be used to solve real-world imperfect-information strategic problems.
[阅读训练day(06)]
