[2312.08484] Q-learners Can Provably Collude in the Iterated Prisoner's Dilemma

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2312.08484] Self-Play Q-learners Can Provably Collude in the Iterated Prisoner's Dilemma Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Computer Science and Game Theory arXiv:2312.08484 (cs) [Submitted on 13 Dec 2023 ( v1 ), last revised 19 Dec 2025 (this version, v3)] Title: Self-Play Q-learners Can Provably Collude in the Iterated Prisoner's Dilemma Authors: Quentin Bertrand , Juan Duque , Emilio Calvano , Gauthier Gidel View a PDF of the paper titled Self-Play Q-learners Can P

related reading

• [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
• [2211.14468] Similarity-based cooperative equilibriumarxiv.org
• [2211.05057] A Note on the Compatibility of Different Robust Program Equilibria of the Prisoner's Dilemmaarxiv.org
• [2512.18552] Toward Training Superintelligent Software Agents through Self-Play SWE-RLarxiv.org
• [2305.18784] Collaborative Multi-Agent Heterogeneous Multi-Armed Banditsarxiv.org
• [1401.5577] Robust Cooperation in the Prisoner's Dilemma: Program Equilibrium via Provability Logicarxiv.org
• [2406.04219] Multi-Agent Imitation Learning: Value is Easy, Regret is Hardarxiv.org
• Reinforcement Learning and Control as Probabilistic Inference: Tutorial and Reviewarxiv.org
• [1903.08894] Towards Characterizing Divergence in Deep Q-Learningarxiv.org
• [2312.12568] Scaling Opponent Shaping to High Dimensional Gamesarxiv.org
• [2512.19428] Attention Is Not What You Needarxiv.org
• [2304.03279] Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmarkarxiv.org