MOMAland: A Set of Benchmarks for Multi-Objective Multi-Agent Reinforcement Learning

TL;DR

MOMAland introduces a comprehensive set of standardized environments and baseline algorithms to facilitate benchmarking and progress in the emerging field of multi-objective multi-agent reinforcement learning.

Contribution

This paper presents MOMAland, the first benchmark suite for multi-objective multi-agent RL, including diverse environments and baseline algorithms for evaluation.

Findings

Over 10 diverse environments included

Baseline algorithms provided for multi-agent multi-objective RL

Supports evaluation of decision-making in complex multi-agent systems

Abstract

Many challenging tasks such as managing traffic systems, electricity grids, or supply chains involve complex decision-making processes that must balance multiple conflicting objectives and coordinate the actions of various independent decision-makers (DMs). One perspective for formalising and addressing such tasks is multi-objective multi-agent reinforcement learning (MOMARL). MOMARL broadens reinforcement learning (RL) to problems with multiple agents each needing to consider multiple objectives in their learning process. In reinforcement learning research, benchmarks are crucial in facilitating progress, evaluation, and reproducibility. The significance of benchmarks is underscored by the existence of numerous benchmark frameworks developed for various RL paradigms, including single-agent RL (e.g., Gymnasium), multi-agent RL (e.g., PettingZoo), and single-agent multi-objective RL (e.g., MO-Gymnasium). To support the advancement of the MOMARL field, we introduce MOMAland, the first collection of standardised environments for multi-objective multi-agent reinforcement learning. MOMAland addresses the need for comprehensive benchmarking in this emerging field, offering over 10 diverse environments that vary in the number of agents, state representations, reward structures, and utility considerations. To provide strong baselines for future research, MOMAland also includes algorithms capable of learning policies in such settings.