Focusing Influence Mechanism for Multi-Agent Reinforcement Learning
Not on the map. Why?
Abstract (arXiv)
Cooperative multi-agent reinforcement learning (MARL) under sparse rewards remains fundamentally challenging because agents often fail to concentrate their influence, leading to insufficiently coordinated exploration. To address this, we propose the Focusing Influence Mechanism (FIM), a framework that encourages agents to focus their influence on under-explored parts of the state space through an entropy-based criterion, while leveraging eligibility traces to enable multiple agents to consistently align and sustain their influence on the same parts of the state space when beneficial, thereby promoting coordinated and persistent joint behavior. By emphasizing under-explored regions of the state space, FIM facilitates more efficient and structured exploration even under extremely sparse rewards. Across diverse MARL benchmarks, FIM consistently improves cooperative performance over strong baselines.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 43 of its 69 references that are in the map. First posted on arXiv 24 Jun 2025.
Builds on
43 of 69 references.
References · 43
- 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
- 2021The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Gamescited by 942
- 2017Counterfactual Multi-Agent Policy Gradientscited by 751
- 2018QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 708
- 2019The StarCraft Multi-Agent Challengecited by 597
- 2019On the Utility of Learning about Humans for Human-AI Coordinationcited by 349
- 2016Safe, Multi-Agent, Reinforcement Learning for Autonomous Drivingcited by 311
- 2020QPLEX: Duplex Dueling Multi-Agent Q-Learningcited by 270
- 2019Google Research Football: A Novel Reinforcement Learning Environmentcited by 260
- 2017Learning with Opponent-Learning Awarenesscited by 222
- 2019MAVEN: Multi-Agent Variational Explorationcited by 185
- 2020RODE: Learning Roles to Decompose Multi-Agent Taskscited by 125
31 more
- 2022SMACv2: An Improved Benchmark for Cooperative Multi-Agent Reinforcement Learningcited by 122
- 2021Celebrating Diversity in Shared Multi-Agent Reinforcement Learningcited by 98
- 2020Learning Latent Representations to Influence Multi-Agent Interactioncited by 93
- 2019Learning Nearly Decomposable Value Functions Via Communication Minimizationcited by 86
- 2019Influence-Based Multi-Agent Explorationcited by 75
- 2019Shapley Q-value: A Local Reward Approach to Solve Global Reward Gamescited by 73
- 2021Rethinking the Implementation Tricks and Monotonicity Constraint in Cooperative Multi-Agent Reinforcement Learningcited by 68
- 2020Learning to Incentivize Other Learning Agentscited by 57
- 2021Cooperative Exploration for Multi-Agent Deep Reinforcement Learningcited by 57
- 2020Learning Individually Inferred Communication for Multi-Agent Cooperationcited by 55
- 2018Stable Opponent Shaping in Differentiable Gamescited by 49
- 2021Episodic Multi-agent Reinforcement Learning with Curiosity-Driven Explorationcited by 49
- 2021Shapley Counterfactual Credits for Multi-Agent Reinforcement Learningcited by 38
- 2023Counterfactual Conservative Q Learning for Offline Multi-agent Reinforcement Learningcited by 37
- 2022MASER: Multi-Agent Reinforcement Learning with Subgoals Generated from Experience Replay Buffercited by 34
- 2019Coordinated Exploration via Intrinsic Rewards for Multi-Agent Reinforcement Learningcited by 30
- 2020Towards Understanding Cooperative Multi-Agent Q-Learning with Value Factorizationcited by 26
- 2021SHAQ: Incorporating Shapley Value Theory into Multi-Agent Q-Learningcited by 26
- 2022PMIC: Improving Multi-Agent Reinforcement Learning with Progressive Mutual Information Collaborationcited by 23
- 2022Influencing Long-Term Behavior in Multiagent Reinforcement Learningcited by 18
- 2022ELIGN: Expectation Alignment as a Multi-Agent Intrinsic Rewardcited by 11
- 2024Understanding Individual Agent Importance in Multi-Agent System via Counterfactual Reasoningcited by 11
- 2021On the Use and Misuse of Absorbing States in Multi-agent Reinforcement Learningcited by 10
- 2023FoX: Formation-aware exploration in multi-agent reinforcement learningcited by 9
- 2023Global Rewards in Multi-Agent Deep Reinforcement Learning for Autonomous Mobility on Demand Systemscited by 8
- 2023Self-Motivated Multi-Agent Explorationcited by 7
- 2024Settling Decentralized Multi-Agent Coordinated Exploration by Novelty Sharingcited by 6
- 2024Imagine, Initialize, and Explore: An Effective Exploration Method in Multi-Agent Reinforcement Learningcited by 5
- 2024Individual Contributions as Intrinsic Exploration Scaffolds for Multi-agent Reinforcement Learningcited by 4
- 2024Reciprocal Reward Influence Encourages Cooperation From Self-Interested Agentscited by 3
- 2023Multi-agent Exploration with Sub-state Entropy Estimationcited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Credit Assignment and Efficient Exploration based on Influence Scope in Multi-agent Reinforcement Learning
- 2025A Principle of Targeted Intervention for Multi-Agent Reinforcement Learning
- 2024Attention-Driven Multi-Agent Reinforcement Learning: Enhancing Decisions with Expertise-Informed Tasks
- 2024MESA: Cooperative Meta-Exploration in Multi-Agent Learning through Exploiting State-Action Space Structure
- 2025MIR: Efficient Exploration in Episodic Multi-Agent Reinforcement Learning via Mutual Intrinsic Reward
- 2024Tacit Learning with Adaptive Information Selection for Cooperative Multi-Agent Reinforcement Learning
- 2025Topology Enhanced MARL for Multi-Agent Cooperative Decision-Making of CAVs
- 2025An Extended Benchmarking of Multi-Agent Reinforcement Learning Algorithms in Complex Fully Cooperative Tasks