Seven Deuce: An Investigation into the Effect of Optimization Techniques in a Head-up No-Limit Poker Agent Using Counterfactual Regret; The GTO Revolution, The Social and Technological Impact of AI on Poker
Xiao, Grant, School of Engineering and Applied Science, University of Virginia
Elliott, Travis, EN-Engineering and Society, University of Virginia
Morrison, Briana, EN-Comp Science Dept, University of Virginia
Carrigan, Coleen, EN-Engineering and Society, University of Virginia
The integration of artificial intelligence into poker presents a compelling case study of how emerging technologies intersect with traditional human practices. This thesis explores that intersection through two complementary projects: a technical paper focused on building an optimized heads-up no-limit poker agent using Counterfactual Regret Minimization and an STS paper examining the ethical and societal implications of AI’s impact on poker. The technical contribution delves into optimization techniques such as game abstraction, hand strength estimation, and branch elimination in order to maximize the efficacy of AI decision making in a complex world. The end aim is to advance reinforcement learning systems capable of operating in high stakes environments, with prospective applications in financial planning and military planning.
In parallel, the STS research investigates how the widespread use of poker AI affects core social values such as fairness, accessibility, and trust within the poker community. Drawing on case studies such as exploitative behavior of the Bot Farm Corporation and the general sentiment from professional players, the paper uncovers how AI tools both empower and destabilize poker as a social and competitive activity. It delves into the changing power dynamics between developers, platforms, and players, and reveals how access to advanced tools can perpetuate inequality.
Collectively, the two papers present a sociotechnical approach that connects engineering and ethics and illustrates how the construction of poker AI needs to be complemented with a critical understanding of its social implications. The two papers aim to emphasize the necessity of coupling technical innovation with ethical consideration to ensure that AI enhances, not displaces, human-oriented values in competitive settings.
BS (Bachelor of Science)
Poker, Artificial Intelligence, Reinforcement Learning
School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Briana Morrison
STS Advisor: Travis Elliott
Technical Team Members: Grant Xiao
English
2025/05/07