Skip to content
arXiv Board & Classic Game AI· Phurinat Polasa·· 2026-10-19AI score28

Teaching AI Agent to Play Poker Strategically Through Preference Optimization

Teaching AI Agent to Play Poker Strategically Through Preference Optimization

AI summary

Study proposes SPOKE, a strategic preference optimization method for training poker agents' strategic decisions, and compares it with SFT, DPO, GRPO, and GSPO. Across three runs with the SmallThinker-4B-A0.6B policy, SPOKE had the lowest average exploitability in Kuhn (0.213) and Leduc (1.081), and the highest top-1 agreement rate on the six-player table (0.775).

Source: arXiv Board & Classic Game AI · arxiv.org