mlsanityData to 7 Oct 2026

CoDreamer: Communication-Based Decentralised World Models

arXiv 2406.13600 ↗First posted 19 Jun 2024cs.AIcited by 5

On the map: World models → World models for language agents

Abstract (arXiv)

Sample efficiency is a critical challenge in reinforcement learning. Model-based RL has emerged as a solution, but its application has largely been confined to single-agent scenarios. In this work, we introduce CoDreamer, an extension of the Dreamer algorithm for multi-agent environments. CoDreamer leverages Graph Neural Networks for a two-level communication system to tackle challenges such as partial observability and inter-agent cooperation. Communication is separately utilised within the learned world models and within the learned policies of each agent to enhance modelling and task-solving. We show that CoDreamer offers greater expressive power than a naive application of Dreamer, and we demonstrate its superiority over baseline methods across various multi-agent environments.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in World models for language agents, one of 11 directions in World models. Builds on 34 of its 49 references that are in the map. First posted on arXiv 19 Jun 2024.

Builds on

34 of 49 references.

References · 34

  1. 2019Dream to Control: Learning Behaviors by Latent Imaginationcited by 1,598
  2. 2019Mastering Atari, Go, Chess and Shogi by Planning with a Learned Modelcited by 1,327
  3. 2018Deeper Insights into Graph Convolutional Networks for Semi-Supervised Learningcited by 1,099
  4. 2020Mastering Atari with Discrete World Modelscited by 990
  5. 2021How Attentive are Graph Attention Networks?cited by 818
  6. 2021Deep Reinforcement Learning at the Edge of the Statistical Precipicecited by 781
  7. 2019The StarCraft Multi-Agent Challengecited by 589
  8. 2016Learning to Communicate with Deep Multi-Agent Reinforcement Learningcited by 561
  9. 2019Quantifying the Carbon Emissions of Machine Learningcited by 516
  10. 2016Learning Multiagent Communication with Backpropagationcited by 355
  11. 2019Google Research Football: A Novel Reinforcement Learning Environmentcited by 257
  12. 2022The Primacy Bias in Deep Reinforcement Learningcited by 242
22 more
  1. 2021Mastering Atari Games with Limited Datacited by 235
  2. 2013The Complexity of Decentralized Control of Markov Decision Processescited by 225
  3. 2020Benchmarking Multi-Agent Deep Reinforcement Learning Algorithms in Cooperative Taskscited by 207
  4. 2021Accounting for Variance in Machine Learning Benchmarkscited by 195
  5. 2018Learning Attentional Communication for Multi-Agent Cooperationcited by 170
  6. 2018Deep Reinforcement Learning for Multi-Agent Systems: A Review of Challenges, Solutions and Applicationscited by 122
  7. 2022SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learningcited by 118
  8. 2017Multiagent Bidirectionally-Coordinated Nets: Emergence of Human-level Coordination in Learning to Play StarCraft Combat Gamescited by 93
  9. 2019Graph Neural Networks for Decentralized Multi-Robot Path Planningcited by 84
  10. 2022VMAS: A Vectorized Multi-Agent Simulator for Collective Robot Learningcited by 63
  11. 2019Learning Decentralized Controllers for Robot Swarms with Graph Neural Networkscited by 62
  12. 2022Towards a Standardised Performance Evaluation Protocol for Cooperative MARLcited by 48
  13. 2022Melting Pot 2.0cited by 37
  14. 2023Heterogeneous Multi-Robot Reinforcement Learningcited by 34
  15. 2020The Emergence of Adversarial Communication in Multi-Agent Reinforcement Learningcited by 28
  16. 2022Controlling Commercial Cooling Systems Using Reinforcement Learningcited by 25
  17. 2022Scalable Multi-Agent Model-Based Reinforcement Learningcited by 24
  18. 2022PrefixRL: Optimization of Parallel Prefix Circuits using Deep Reinforcement Learningcited by 18
  19. 2021MAMBPO: Sample-efficient multi-robot reinforcement learning using learned world modelscited by 16
  20. 2022QGNN: Value Function Factorisation with Graph Neural Networkscited by 14
  21. 2017Revisiting the Master-Slave Architecture in Multi-Agent Deep Reinforcement Learningcited by 11
  22. 2018Graph Neural Networks for Learning Robot Team Coordinationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 2; 2026: 2.

5 papers.

  1. 2024Decentralized Transformers with Centralized Aggregation are Sample-Efficient Multi-Agent World Modelscited by 10
  2. 2025Empowering Multi-Robot Cooperation via Sequential World Modelscited by 9
  3. 2025In-Context Reinforcement Learning via Communicative World Modelscited by 3
  4. 2026Puzzle it Out: Local-to-Global World Model for Offline Multi-Agent Reinforcement Learningcited by 2
  5. 2026Dreaming Of Others: Latent Teammate Modeling In World Models For Multi-Agent Reinforcement Learningcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Dreamer-CPC: Message Learning with World Models for Decentralized Multi-agent Reinforcement Learning
  2. 2025Communicating Plans, Not Percepts: Scalable Multi-Agent Coordination with Embodied World Models
  3. 2025Decentralized Collective World Model for Emergent Communication and Coordination
  4. 2026The Five Ws of Multi-Agent Communication: Who Talks to Whom, When, What, and Why -- A Survey from MARL to Emergent Language and LLMs
  5. 2025Transformer World Model for Sample Efficient Multi-Agent Reinforcement Learning
  6. 2026Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
  7. 2025LLM-Powered Decentralized Generative Agents with Adaptive Hierarchical Knowledge Graph for Cooperative Planning
  8. 2025Enhancing Multi-Agent Systems via Reinforcement Learning with LLM-based Planner and Graph-based Policy