FCMNet: Full Communication Memory Net for Team-Level Cooperation in Multi-Agent Systems
Yutong Wang, Guillaume Sartoretti

TL;DR
FCMNet is a reinforcement learning approach enabling multi-agent systems to learn effective multi-hop communication protocols and decentralized policies, improving cooperation in partially observable environments with unreliable communication.
Contribution
Introduces FCMNet, a novel method that learns multi-hop communication and team policies simultaneously using hidden states of directional RNNs, handling unreliable communication channels.
Findings
Outperforms state-of-the-art methods in StarCraft II micromanagement tasks.
Shows robustness under message loss and non-differentiable communication channels.
Effective in collaborative multi-agent pathfinding with shared and individual rewards.
Abstract
Decentralized cooperation in partially-observable multi-agent systems requires effective communications among agents. To support this effort, this work focuses on the class of problems where global communications are available but may be unreliable, thus precluding differentiable communication learning methods. We introduce FCMNet, a reinforcement learning based approach that allows agents to simultaneously learn a) an effective multi-hop communications protocol and b) a common, decentralized policy that enables team-level decision-making. Specifically, our proposed method utilizes the hidden states of multiple directional recurrent neural networks as communication messages among agents. Using a simple multi-hop topology, we endow each agent with the ability to receive information sequentially encoded by every other agent at each time step, leading to improved global cooperation. We…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsReinforcement Learning in Robotics · Neural Networks and Reservoir Computing · Advanced Memory and Neural Computing
