# RGMComm: Return Gap Minimization via Discrete Communications in Multi-Agent Reinforcement Learning

arXiv: https://arxiv.org/abs/2308.03358 · HTML: https://mlsanity.com/paper/2308.03358/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 7 Aug 2023.

## Builds on (most cited first, 18 of 18)

- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments](https://mlsanity.com/paper/1706.02275/) (2017)
- [The Concrete Distribution: A Continuous Relaxation of Discrete Random Variables](https://mlsanity.com/paper/1611.00712/) (2016)
- [QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1803.11485/) (2018)
- [Learning to Communicate with Deep Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/1605.06676/) (2016)
- [Learning Multiagent Communication with Backpropagation](https://mlsanity.com/paper/1605.07736/) (2016)
- [Emergence of Grounded Compositional Language in Multi-Agent Populations](https://mlsanity.com/paper/1703.04908/) (2017)
- [TarMAC: Targeted Multi-Agent Communication](https://mlsanity.com/paper/1810.11187/) (2018)
- [Learning Attentional Communication for Multi-Agent Cooperation](https://mlsanity.com/paper/1805.07733/) (2018)
- [Emergent Multi-Agent Communication in the Deep Learning Era](https://mlsanity.com/paper/2006.02419/) (2020)
- [Learning Discrete Representations via Information Maximizing Self-Augmented Training](https://mlsanity.com/paper/1702.08720/) (2017)
- [Emergence of Language with Multi-agent Games: Learning to Communicate with Sequences of Symbols](https://mlsanity.com/paper/1705.11192/) (2017)
- [Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Tasks](https://mlsanity.com/paper/1812.09755/) (2018)
- [Off-Policy Multi-Agent Decomposed Policy Gradients](https://mlsanity.com/paper/2007.12322/) (2020)
- [Learning Nearly Decomposable Value Functions Via Communication Minimization](https://mlsanity.com/paper/1910.05366/) (2019)
- [Emergent Discrete Communication in Semantic Spaces](https://mlsanity.com/paper/2108.01828/) (2021)
- [MAC-PO: Multi-Agent Experience Replay via Collective Priority Optimization](https://mlsanity.com/paper/2302.10418/) (2023)
- [Learning Emergent Discrete Message Communication for Cooperative Reinforcement Learning](https://mlsanity.com/paper/2102.12550/) (2021)

## Built on by (most cited first, 8 of 8)

- [ImageFlowNet: Forecasting Multiscale Image-Level Trajectories of Disease Progression with Irregularly-Sampled Longitudinal Medical Images](https://mlsanity.com/paper/2406.14794/) (2024)
- [BadCLM: Backdoor Attack in Clinical Language Models for Electronic Health Records](https://mlsanity.com/paper/2407.05213/) (2024)
- [Noise Distribution Decomposition based Multi-Agent Distributional Reinforcement Learning](https://mlsanity.com/paper/2312.07025/) (2023)
- [RGMDT: Return-Gap-Minimizing Decision Tree Extraction in Non-Euclidean Metric Space](https://mlsanity.com/paper/2410.16517/) (2024)
- [Statistically Efficient Variance Reduction with Double Policy Estimation for Off-Policy Evaluation in Sequence-Modeled Reinforcement Learning](https://mlsanity.com/paper/2308.14897/) (2023)
- [DCMAC: Demand-aware Customized Multi-Agent Communication via Upper Bound Training](https://mlsanity.com/paper/2409.07127/) (2024)
- [Hierarchical Multi-agent Meta-Reinforcement Learning for Cross-channel Bidding](https://mlsanity.com/paper/2412.19064/) (2024)
- [MUTE: Return-Preserving Communication Unlearning for Efficient Multi-Agent Coordination](https://mlsanity.com/paper/2607.03473/) (2026)

## Abstract (arXiv)

Communication is crucial for solving cooperative Multi-Agent Reinforcement Learning tasks in partially observable Markov Decision Processes. Existing works often rely on black-box methods to encode local information/features into messages shared with other agents, leading to the generation of continuous messages with high communication overhead and poor interpretability. Prior attempts at discrete communication methods generate one-hot vectors trained as part of agents' actions and use the Gumbel softmax operation for calculating message gradients, which are all heuristic designs that do not provide any quantitative guarantees on the expected return. This paper establishes an upper bound on the return gap between an ideal policy with full observability and an optimal partially observable policy with discrete communication. This result enables us to recast multi-agent communication into a novel online clustering problem over the local observations at each agent, with messages as cluster labels and the upper bound on the return gap as clustering loss. To minimize the return gap, we propose the Return-Gap-Minimization Communication (RGMComm) algorithm, which is a surprisingly simple design of discrete message generation functions and is integrated with reinforcement learning through the utilization of a novel Regularized Information Maximization loss function, which incorporates cosine-distance as the clustering metric. Evaluations show that RGMComm significantly outperforms state-of-the-art multi-agent communication baselines and can achieve nearly optimal returns with few-bit messages that are naturally interpretable.