RL-CFR: Improving Action Abstraction for Imperfect Information Extensive-Form Games with Reinforcement Learning
Li, Boning, Fang, Zhixuan, Huang, Longbo
–arXiv.org Artificial Intelligence
Effective action abstraction is crucial in tackling challenges associated with large action spaces in Imperfect Information Extensive-Form Games (IIEFGs). However, due to the vast state space and computational complexity in IIEFGs, existing methods often rely on fixed abstractions, resulting in sub-optimal performance. In response, we introduce RL-CFR, a novel reinforcement learning (RL) approach for dynamic action abstraction. RL-CFR builds upon our innovative Markov Decision Process (MDP) formulation, with states corresponding to public information and actions represented as feature vectors indicating specific action abstractions. The reward is defined as the expected payoff difference between the selected and default action abstractions. RL-CFR constructs a game tree with RL-guided action abstractions and utilizes counterfactual regret minimization (CFR) for strategy derivation. Impressively, it can be trained from scratch, achieving higher expected payoff without increased CFR solving time. In experiments on Heads-up No-limit Texas Hold'em, RL-CFR outperforms ReBeL's replication and Slumbot, demonstrating significant win-rate margins of $64\pm 11$ and $84\pm 17$ mbb/hand, respectively.
arXiv.org Artificial Intelligence
Mar-7-2024
- Country:
- Africa > Rwanda
- Asia
- China > Beijing
- Beijing (0.04)
- Middle East > Republic of Türkiye
- Istanbul Province > Istanbul (0.04)
- China > Beijing
- Europe
- Hungary > Budapest
- Budapest (0.04)
- Middle East > Republic of Türkiye
- Istanbul Province > Istanbul (0.04)
- Spain > Catalonia
- Barcelona Province > Barcelona (0.04)
- United Kingdom
- England
- Cambridgeshire > Cambridge (0.28)
- East Sussex (0.04)
- Greater London > London (0.04)
- Scotland (0.04)
- England
- Hungary > Budapest
- North America
- Canada
- Alberta > Census Division No. 11
- Edmonton Metropolitan Region > Edmonton (0.04)
- British Columbia > Metro Vancouver Regional District
- Vancouver (0.04)
- Ontario > Toronto (0.04)
- Quebec
- Capitale-Nationale Region
- Quebec City (0.04)
- Québec (0.04)
- Montreal (0.04)
- Capitale-Nationale Region
- Alberta > Census Division No. 11
- Puerto Rico > San Juan
- San Juan (0.04)
- United States
- Arizona > Maricopa County
- Phoenix (0.04)
- California
- Los Angeles County
- Long Beach (0.14)
- Pasadena (0.04)
- San Diego County > San Diego (0.04)
- San Francisco County > San Francisco (0.14)
- Santa Clara County > Stanford (0.04)
- Los Angeles County
- Colorado > Denver County
- Denver (0.04)
- District of Columbia > Washington (0.04)
- Hawaii > Honolulu County
- Honolulu (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Minnesota > Ramsey County
- Saint Paul (0.04)
- Texas > Travis County
- Austin (0.04)
- Arizona > Maricopa County
- Canada
- Oceania
- Australia > Victoria
- Melbourne (0.04)
- New Zealand > North Island
- Auckland Region > Auckland (0.04)
- Australia > Victoria
- Genre:
- Research Report > New Finding (0.93)
- Industry:
- Leisure & Entertainment > Games > Poker (0.49)
- Technology: