|
Briefing
|
- Poker Bot– Researchers in Canada and Czech Republic developed poker playing bot DeepStack that won against professional human players in no-limit Texas hold’em poker
- Deep Learning Program – Learned by playing against itself, refining its strategy after each game, with researchers feeding system with previous poker games
- Gut Feel Emulation – Researchers applied approximation technique on DeepStack to emulate human gut feel, which is used for guessing if other players are bluffing or not, enabling it to study opponents' betting patterns, rather than body language
- Significantly Higher Win Rate – Recorded win rate is nine times better than professional player performance
|
|
Accelerator
|
|
|
Sector
|
Information Technology
|
|
Organization
|
Charles University, Czech Technical University in Prague, University of Alberta
|
|
Source
|
|
|
Original Publication Date
|
January 11, 2017
|
