arXiv Board & Classic Game AI· Boning Li·· 3 d agoPickAI score56
Paper proposes conditional safety theorem and a posteriori certificate for subgame solving in multiplayer games
Safe Subgame Resolving in Multiplayer Games
AI summary
The paper proposes a conditional safety theorem for subgame solving in multiplayer games with three or more players acting independently, decomposes the best-response gap upper bound of the post-solve strategy into four terms, and gives an a posteriori certificate that can check any submitted update.
Why it matters
The certificate requires only one best-response computation per opponent, making it usable for online safety verification of real-time multiplayer solvers.
Source: arXiv Board & Classic Game AI · arxiv.org