Communication in Multi-Agent RL Reading List

Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026

Communication lets decentralized agents coordinate when observations and rewards alone are not enough. These papers show how differentiable channels, attention, and scheduling turn cheap talk into useful protocols.

Communication in Multi-Agent RL: 10 key papers

1. Emergence of Grounded Compositional Language in Multi-Agent Populations
   Mordatch and Abbeel. arXiv 2017.
2. Learning Attentional Communication for Multi-Agent Cooperation
   Jiang and Lu. arXiv 2018.
3. TarMAC: Targeted Multi-Agent Communication
   Das et al. arXiv 2018.
4. Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Tasks
   Singh et al. arXiv 2018.
5. Biases for Emergent Communication in Multi-agent Reinforcement Learning
   Eccles et al. arXiv 2019.
6. Learning Multiagent Communication with Backpropagation
   Sukhbaatar et al. arXiv 2016.
7. Learning to Communicate with Deep Multi-Agent Reinforcement Learning
   Foerster et al. arXiv 2016.
8. Multi-Agent Cooperation and the Emergence of (Natural) Language
   Lazaridou et al. arXiv 2016.
9. Emergent Communication through Negotiation
   Cao et al. arXiv 2018.
10. Learning Efficient Multi-agent Communication: An Information Bottleneck Approach
    Wang et al. arXiv 2019.

Related reading lists

• Emergent Behavior & Self-Organization
• Multi-Agent Reinforcement Learning
• Swarm RL & Collective Behavior
• Information Theory & RL
• Game Theory for Multi-Agent Systems
• Federated & Distributed Learning
• Scaling Laws and Emergent Capabilities
• Graph Neural Networks & Message Passing