Heterogeneity in Multi-Agent Reinforcement Learning
Preprint
Abstract
Heterogeneity is a fundamental property in multi-agent reinforcement learning (MARL), which is closely related not only to the functional differences of agents, but also to policy diversity and environmental interactions. However, the MARL field currently lacks a rigorous definition and deeper understanding of heterogeneity. This paper systematically discusses heterogeneity in MARL from the perspectives of definition, quantification, and utilization. First, based on an agent-level modeling of MARL, we categorize heterogeneity into five types and provide mathematical definitions. Second, we define the concept of heterogeneity distance and propose a practical quantification method. Third, we design a heterogeneity-based multi-agent dynamic parameter sharing algorithm as an example of the application of our methodology. Case studies demonstrate that our method can effectively identify and quantify various types of agent heterogeneity. Experimental results show that the proposed algorithm, compared to other parameter sharing baselines, has better interpretability and stronger adaptability. The proposed methodology will help the MARL community gain a more comprehensive and profound understanding of heterogeneity, and further promote the development of practical algorithms.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- Heterogeneous multi-agent reinforcement learning for zero-shot scalable collaborationNeurocomputing, 2025
- A survey on multi-agent reinforcement learning and its applicationJournal of Automation and Intelligence, 2024
- GHQ: grouped hybrid Q-learning for cooperative heterogeneous multi-agent reinforcement learningComplex & Intelligent Systems, 2024
- Toward multi-target self-organizing pursuit in a partially observable Markov gameInformation Sciences, 2023
- Conditional Variational Autoencoder for Learned Image ReconstructionComputation, 2021
- Behavioural interdependence in a shrimp‐goby mutualismJournal of Zoology, 2019
- Conditional Variational Autoencoder for Prediction and Feature Recovery Applied to Intrusion Detection in IoTSensors, 2017
- Distributed Optimal Power Flow for Smart MicrogridsIEEE Transactions on Smart Grid, 2013
- Cooperative Multi-Agent Learning: The State of the ArtAutonomous Agents and Multi-Agent Systems, 2005
- Lifelong Adaptation in Heterogeneous Multi-Robot Teams: Response to Continual Variation in Individual Robot PerformanceAutonomous Robots, 2000