|国家预印本平台
首页|MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning

MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning

MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning

来源:Arxiv_logoArxiv
英文摘要

Benchmarks play a crucial role in the development and analysis of reinforcement learning (RL) algorithms, with environment availability strongly impacting research. One particularly underexplored intersection is continual learning (CL) in cooperative multi-agent settings. To remedy this, we introduce MEAL (Multi-agent Environments for Adaptive Learning), the first benchmark tailored for continual multi-agent reinforcement learning (CMARL). Existing CL benchmarks run environments on the CPU, leading to computational bottlenecks and limiting the length of task sequences. MEAL leverages JAX for GPU acceleration, enabling continual learning across sequences of 100 tasks on a standard desktop PC in a few hours. We show that naively combining popular CL and MARL methods yields strong performance on simple environments, but fails to scale to more complex settings requiring sustained coordination and adaptation. Our ablation study identifies architectural and algorithmic features critical for CMARL on MEAL.

Tristan Tomilin、Luka van den Boogaard、Samuel Garcin、Bram Grooten、Meng Fang、Mykola Pechenizkiy

计算技术、计算机技术

Tristan Tomilin,Luka van den Boogaard,Samuel Garcin,Bram Grooten,Meng Fang,Mykola Pechenizkiy.MEAL: A Benchmark for Continual Multi-Agent Reinforcement Learning[EB/OL].(2025-06-17)[2025-07-20].https://arxiv.org/abs/2506.14990.点此复制

评论