mlsanityData to 7 Oct 2026

Learning Selective Communication for Multi-Agent Path Finding

arXiv 2109.05413 ↗First posted 12 Sep 2021cs.RO cs.AI cs.MAcited by 32

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Learning communication via deep reinforcement learning (RL) or imitation learning (IL) has recently been shown to be an effective way to solve Multi-Agent Path Finding (MAPF). However, existing communication based MAPF solvers focus on broadcast communication, where an agent broadcasts its message to all other or predefined agents. It is not only impractical but also leads to redundant information that could even impair the multi-agent cooperation. A succinct communication scheme should learn which information is relevant and influential to each agent's decision making process. To address this problem, we consider a request-reply scenario and propose Decision Causal Communication (DCC), a simple yet efficient model to enable agents to select neighbors to conduct communication during both training and execution. Specifically, a neighbor is determined as relevant and influential only when the presence of this neighbor causes the decision adjustment on the central agent. This judgment is learned only based on agent's local observation and thus suitable for decentralized execution to handle large scale problems. Empirical evaluation in obstacle-rich environment indicates the high success rate with low communication overhead of our method.

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 25 of its 43 references that are in the map. First posted on arXiv 12 Sep 2021.

Builds on

25 of 43 references.

References · 25

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,341
  3. 2016Asynchronous Methods for Deep Reinforcement Learningcited by 2,218
  4. 2015Dueling Network Architectures for Deep Reinforcement Learningcited by 769
  5. 2018QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 697
  6. 2017Ray: A Distributed Framework for Emerging AI Applicationscited by 516
  7. 2018Distributed Prioritized Experience Replaycited by 260
  8. 2018TarMAC: Targeted Multi-Agent Communicationcited by 201
  9. 2019Multi-Agent Pathfinding: Definitions, Variants, and Benchmarkscited by 197
  10. 2018Graph Convolutional Reinforcement Learningcited by 148
  11. 2018Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Taskscited by 114
  12. 2018Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learningcited by 101
13 more
  1. 2018PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learningcited by 89
  2. 2019Learning to Schedule Communication in Multi-agent Reinforcement Learningcited by 89
  3. 2019Learning Nearly Decomposable Value Functions Via Communication Minimizationcited by 85
  4. 2019Graph Neural Networks for Decentralized Multi-Robot Path Planningcited by 84
  5. 2020PRIMAL2: Pathfinding via Reinforcement and Imitation Multi-Agent Learning -- Lifelongcited by 61
  6. 2020Learning Individually Inferred Communication for Multi-Agent Cooperationcited by 55
  7. 2019Efficient Communication in Multi-Agent Reinforcement Learning via Variance Based Controlcited by 46
  8. 2020Message-Aware Graph Attention Networks for Large-Scale Multi-Robot Path Planningcited by 41
  9. 2020Mobile Robot Path Planning in Dynamic Environments through Globally Guided Reinforcement Learningcited by 36
  10. 2021Distributed Heuristic Multi-Agent Path Finding with Communicationcited by 31
  11. 2020Succinct and Robust Multi-Agent Communication With Temporal Message Controlcited by 29
  12. 2020MAPPER: Multi-Agent Path Planning with Evolutionary Reinforcement Learning in Mixed Dynamic Environmentscited by 26
  13. 2020GLAS: Global-to-Local Safe Autonomy Synthesis for Multi-Robot Motion Planning with End-to-End Learningcited by 21

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 6; 2024: 10; 2025: 10; 2026: 5.

32 papers.

Sort:

By area:

  1. 2024POGEMA: A Benchmark Platform for Cooperative Multi-Agent Pathfindingcited by 20
  2. 2024MAPF-GPT: Imitation Learning for Multi-Agent Pathfinding at Scalecited by 17
  3. 2023Vehicle-to-Everything Cooperative Perception for Autonomous Drivingcited by 15
  4. 2022Distributed Reinforcement Learning for Robot Teams: A Reviewcited by 11
  5. 2024Work Smarter Not Harder: Simple Imitation Learning with CS-PIBT Outperforms Large Scale Imitation Learning for MAPFcited by 10
  6. 2024Towards Generalizability of Multi-Agent Reinforcement Learning in Graphs with Recurrent Message Passingcited by 8
  7. 2025SRMT: Shared Memory for Multi-agent Lifelong Pathfindingcited by 8
  8. 2024Ensembling Prioritized Hybrid Policies for Multi-agent Pathfindingcited by 7
  9. 2023SACHA: Soft Actor-Critic with Heuristic-Based Attention for Partially Observable Multi-Agent Path Findingcited by 6
  10. 2024Deploying Ten Thousand Robots: Scalable Imitation Learning for Lifelong Multi-Agent Path Findingcited by 6
  11. 2023Selective Communication for Cooperative Perception in End-to-End Autonomous Drivingcited by 5
  12. 2023Learning Team-Based Navigation: A Review of Deep Reinforcement Learning Techniques for Multi-Agent Pathfindingcited by 5
20 more
  1. 2025RAILGUN: A Unified Convolutional Policy for Multi-Agent Path Finding Across Different Environments and Taskscited by 5
  2. 2025Advancing Learnable Multi-Agent Pathfinding Solvers with Active Fine-Tuningcited by 5
  3. 2025Graph Attention-Guided Search for Dense Multi-Agent Pathfindingcited by 5
  4. 2023Robust Communicative Multi-Agent Reinforcement Learning with Active Defensecited by 4
  5. 2024HiMAP: Learning Heuristics-Informed Policies for Large-Scale Multi-Agent Pathfindingcited by 4
  6. 2025SIGMA: Sheaf-Informed Geometric Multi-Agent Pathfindingcited by 4
  7. 2026Pairwise is Not Enough: Hypergraph Neural Networks for Multi-Agent Pathfindingcited by 3
  8. 2025SkyRover: A Modular Simulator for Cross-Domain Pathfindingcited by 1
  9. 2025LLMDR: LLM-Driven Deadlock Detection and Resolution in Multi-Agent Pathfindingcited by 1
  10. 2026Learning to Communicate Locally for Large-Scale Multi-Agent Pathfindingcited by 1
  11. 2023Low Entropy Communication in Multi-Agent Reinforcement Learningcited by 0
  12. 2024Cooperative Reward Shaping for Multi-Agent Pathfindingcited by 0
  13. 2024Improving Zero-Shot ObjectNav with Generative Communicationcited by 0
  14. 2024MFC-EQ: Mean-Field Control with Envelope Q-Learning for Moving Decentralized Agents in Formationcited by 0
  15. 2025Multi-Agent Path Finding via Offline RL and LLM Collaborationcited by 0
  16. 2025Sequence Pathfinder for Multi-Agent Pickup and Delivery in the Warehousecited by 0
  17. 2025Towards Information-Optimized Multi-Agent Path Finding: A Hybrid Framework with Reduced Inter-Agent Information Sharingcited by 0
  18. 2026Search-Aided Joint Agent-Environment Reinforcement Learning for Robust Lifelong Multi-Agent Path Finding with Rotationscited by 0
  19. 2026Decentralized Master-Mind: Joint Action Refinement through Iterative Intent Denoising in Multi-Agent Pathfindingcited by 0
  20. 2026Managing Context and Communication in Distributed Agentic UAV Swarmscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026The Five Ws of Multi-Agent Communication: Who Talks to Whom, When, What, and Why -- A Survey from MARL to Emergent Language and LLMs
  2. 2025Communicating Plans, Not Percepts: Scalable Multi-Agent Coordination with Embodied World Models
  3. 2023Efficient Communication via Self-supervised Information Aggregation for Online and Offline Multi-agent Reinforcement Learning
  4. 2026LLM-Guided Communication for Cooperative Multi-Agent Reinforcement Learning
  5. 2026Agent Q-Mix: Selecting the Right Action for LLM Multi-Agent Systems through Reinforcement Learning
  6. 2026Learning Communication-Conditioned Generative Policies for Decentralized Multi-Agent Collision Avoidance ↗
  7. 2026Dynamic Coalition Formation and Communication Pricing in Skill-Based Agentic AI Systems
  8. 2019Learning to Request Guidance in Emergent Communication