mlsanityData to 7 Oct 2026

ROMA: Multi-Agent Reinforcement Learning with Emergent Roles

arXiv 2003.08039 ↗First posted 18 Mar 2020cs.MAcited by 135

Not on the map. Why?

Abstract (arXiv)

The role concept provides a useful tool to design and understand complex multi-agent systems, which allows agents with a similar role to share similar behaviors. However, existing role-based methods use prior domain knowledge and predefine role structures and behaviors. In contrast, multi-agent reinforcement learning (MARL) provides flexibility and adaptability, but less efficiency in complex tasks. In this paper, we synergize these two paradigms and propose a role-oriented MARL framework (ROMA). In this framework, roles are emergent, and agents with similar roles tend to share their learning and to be specialized on certain sub-tasks. To this end, we construct a stochastic role embedding space by introducing two novel regularizers and conditioning individual policies on roles. Experiments show that our method can learn specialized, dynamic, and identifiable roles, which help our method push forward the state of the art on the StarCraft II micromanagement benchmark. Demonstrative videos are available at https://sites.google.com/view/romarl/.

Key facts. Cited by 135 papers in the map (data through 7 Oct 2026). First posted on arXiv 18 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 21; 2022: 19; 2023: 24; 2024: 29; 2025: 24; 2026: 18.

135 papers.

Sort:

By area:

  1. 2023Human-Timescale Adaptation in an Open-Ended Task Spacecited by 136
  2. 2022SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learningcited by 118
  3. 2021Celebrating Diversity in Shared Multi-Agent Reinforcement Learningcited by 96
  4. 2021Scaling Multi-Agent Reinforcement Learning with Selective Parameter Sharingcited by 85
  5. 2021Contrasting Centralized and Decentralized Critics in Multi-Agent Reinforcement Learningcited by 54
  6. 2025Vision-Zero: Scalable VLM Self-Improvement via Strategic Gamified Self-Playcited by 37
  7. 2022MASER: Multi-Agent Reinforcement Learning with Subgoals Generated from Experience Replay Buffercited by 34
  8. 2023Heterogeneous Multi-Robot Reinforcement Learningcited by 34
  9. 2023Dynamic Handover: Throw and Catch with Bimanual Handscited by 33
  10. 2021Coach-Player Multi-Agent Reinforcement Learning for Dynamic Team Compositioncited by 28
  11. 2021DFAC Framework: Factorizing the Value Function via Quantile Mixture for Multi-Agent Distributional Q-Learningcited by 26
  12. 2024Group-Aware Coordination Graph for Multi-Agent Reinforcement Learningcited by 17
38 more
  1. 2021Causal Multi-Agent Reinforcement Learning: Review and Open Problemscited by 16
  2. 2022LDSA: Learning Dynamic Subtask Assignment in Cooperative Multi-Agent Reinforcement Learningcited by 15
  3. 2021Self-Organized Polynomial-Time Coordination Graphscited by 14
  4. 2022Multi-Agent Policy Transfer via Task Relationship Modelingcited by 14
  5. 2024GemNet: Menu-Based, Strategy-Proof Multi-Bidder Auctions Through Deep Learningcited by 14
  6. 2021Cooperative Multi-Agent Transfer Learning with Level-Adaptive Credit Assignmentcited by 12
  7. 2021Adaptable Agent Populations via a Generative Model of Policiescited by 12
  8. 2023Effective and Stable Role-Based Multi-Agent Collaboration by Structural Information Principlescited by 12
  9. 2023System Neural Diversity: Measuring Behavioral Heterogeneity in Multi-Agent Learningcited by 12
  10. 2022Policy Diagnosis via Measuring Role Diversity in Cooperative Multi-agent RLcited by 11
  11. 2022Non-Linear Coordination Graphscited by 10
  12. 2024Controlling Behavioral Diversity in Multi-Agent Reinforcement Learningcited by 10
  13. 2024Towards Collaborative Intelligence: Propagating Intentions and Reasoning for Multi-Agent Coordination with Large Language Modelscited by 10
  14. 2024HyperMARL: Adaptive Hypernetworks for Multi-Agent RLcited by 10
  15. 2021LINDA: Multi-Agent Local Information Decomposition for Awareness of Teammatescited by 9
  16. 2023FoX: Formation-aware exploration in multi-agent reinforcement learningcited by 9
  17. 2024Multi-Sender Persuasion: A Computational Perspectivecited by 9
  18. 2024On Centralized Critics in Multi-Agent Reinforcement Learningcited by 9
  19. 2024Towards Foundation-model-based Multiagent System to Accelerate AI for Social Impactcited by 9
  20. 2022Sparse Adversarial Attack in Multi-agent Reinforcement Learningcited by 8
  21. 2022Consensus Learning for Cooperative Multi-Agent Reinforcement Learningcited by 8
  22. 2022Interaction Pattern Disentangling for Multi-Agent Reinforcement Learningcited by 8
  23. 2023Semantically Aligned Task Decomposition in Multi-Agent Reinforcement Learningcited by 8
  24. 2025A Taxonomy of Hierarchical Multi-Agent Systems: Design Patterns, Coordination Mechanisms, and Industrial Applicationscited by 8
  25. 2021SIDE: State Inference for Partially Observable Cooperative Multi-Agent Reinforcement Learningcited by 7
  26. 2021Learning Meta Representations for Agents in Multi-Agent Reinforcement Learningcited by 6
  27. 2022Mingling Foresight with Imagination: Model-Based Cooperative Multi-Agent Reinforcement Learningcited by 6
  28. 2024Heterogeneous Multi-Agent Reinforcement Learning for Zero-Shot Scalable Collaborationcited by 6
  29. 2024Variational Offline Multi-agent Skill Discoverycited by 6
  30. 2024Grounded Answers for Multi-agent Decision-making Problem through Generative World Modelcited by 6
  31. 2021Provably Efficient Multi-Agent Reinforcement Learning with Fully Decentralized Communicationcited by 5
  32. 2022Learning to Transfer Role Assignment Across Team Sizescited by 5
  33. 2022Multi-Agent Reinforcement Learning with Shared Resources for Inventory Managementcited by 5
  34. 2023STAS: Spatial-Temporal Return Decomposition for Multi-agent Reinforcement Learningcited by 5
  35. 2024Inferring Latent Temporal Sparse Coordination Graph for Multi-Agent Reinforcement Learningcited by 5
  36. 2024The Bandit Whisperer: Communication Learning for Restless Banditscited by 5
  37. 2024Kaleidoscope: Learnable Masks for Heterogeneous Multi-agent Reinforcement Learningcited by 5
  38. 2024On Diffusion Models for Multi-Agent Partial Observability: Shared Attractors, Error Bounds, and Composite Flowcited by 5

50 most cited of 135 All 135 (JSON)