Sparse reward environments pose significant challenges in reinforcement learning, especially within multi-agent systems (MAS) where feedback is delayed and shared across agents, leading to suboptimal learning. We propose Collaborative Multi-dimensional Course Learning (CCL), a novel curriculum learning framework that addresses this by (1) refining intermediate tasks for individual agents, (2) using a variational evolutionary algorithm to generate informative subtasks, and (3) co-evolving agents with their environment to enhance training stability. Experiments on five cooperative tasks in the MPE and Hide-and-Seek environments show that CCL outperforms existing methods in sparse reward settings.
@article{arxiv.2505.07854,
title = {CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution},
author = {Yufei Lin and Chengwei Ye and Huanzhen Zhang and Kangsheng Wang and Linuo Xu and Shuyan Liu and Zeyu Zhang},
journal= {arXiv preprint arXiv:2505.07854},
year = {2025}
}