Chapter
The Success of AI in Six-Player Poker
59:59 - 1:08:22 (08:23)
The techniques used in two-player poker to approximate an equilibrium still work in practice for six-player poker despite not offering guarantees outside of two-player zero-sum games, leading to debate among academics and the poker community.
Clips
A poker bot was created by Noam Brown and his team, named 'Pluribus' that beat humans at a variety of poker games.
59:59 - 1:01:05 (01:05)
Summary
A poker bot was created by Noam Brown and his team, named 'Pluribus' that beat humans at a variety of poker games. By the end, the bot had won for three days straight and proved to be the better player than its human opponents.
ChapterThe Success of AI in Six-Player Poker
Episode#344 – Noam Brown: AI vs Humans in Poker and Games of Strategic Negotiation
PodcastLex Fridman Podcast
The extension of AI techniques to six-player poker games was debated amongst the academic and poker communities due to the complexity of the game, as the exhaustive search method used for two-player games is no longer effective in larger games.
1:01:05 - 1:04:54 (03:49)
Summary
The extension of AI techniques to six-player poker games was debated amongst the academic and poker communities due to the complexity of the game, as the exhaustive search method used for two-player games is no longer effective in larger games.
ChapterThe Success of AI in Six-Player Poker
Episode#344 – Noam Brown: AI vs Humans in Poker and Games of Strategic Negotiation
PodcastLex Fridman Podcast
The techniques used in two-player poker to approximate an equilibrium strategy have been found to work well in six-player poker as well, due to its adversarial nature where players don't collaborate, suggestions are made that self-play may potentially work for approximations.
1:04:54 - 1:08:22 (03:28)
Summary
The techniques used in two-player poker to approximate an equilibrium strategy have been found to work well in six-player poker as well, due to its adversarial nature where players don't collaborate, suggestions are made that self-play may potentially work for approximations. However, computing an actual Nash equilibrium for every single player, there is no guarantee that the joint strategy will be a Nash equilibrium.