References
These references provide the formal background and research context used throughout the course. The concept pages introduce each idea before assigning outside reading, while the research connections explain why a selected paper matters for coordination, communication, or adaptation. Start with the two textbooks for broad coverage, then use the primary papers to examine benchmark design, interpretable communication, ad hoc teamwork, and unseen-partner evaluation in greater depth.
Foundations
Section titled “Foundations”- Richard S. Sutton and Andrew G. Barto. Reinforcement Learning: An Introduction. Second edition, MIT Press, 2018.
- Stefano V. Albrecht, Filippos Christianos, and Lukas Schäfer. Multi-Agent Reinforcement Learning: Foundations and Modern Approaches. MIT Press, 2024. Open textbook.
- Kevin P. Murphy. Reinforcement Learning: A Comprehensive Overview, 2025. arXiv:2412.05265.
Coordination and Evaluation
Section titled “Coordination and Evaluation”- Mikayel Samvelyan et al. The StarCraft Multi-Agent Challenge, 2019. arXiv:1902.04043.
- Benjamin Ellis et al. SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learning. NeurIPS 2023 Datasets and Benchmarks Track. arXiv:2212.07489.
Communication
Section titled “Communication”- Huao Li et al. Language Grounded Multi-agent Reinforcement Learning with Human-interpretable Communication. NeurIPS 2024. Proceedings.
Adaptation and Unseen Partners
Section titled “Adaptation and Unseen Partners”- Caroline Wang et al. N-Agent Ad Hoc Teamwork. NeurIPS 2024. Proceedings.
- Xihuai Wang et al. ZSC-Eval: An Evaluation Toolkit and Benchmark for Multi-agent Zero-shot Coordination. NeurIPS 2024 Datasets and Benchmarks Track. Proceedings.