通过协同演化设计任务，提升稀疏奖励下多智能体学习效果
CCL: Collaborative Curriculum Learning for Sparse-Reward Multi-Agent Reinforcement Learning via Co-evolutionary Task Evolution
- 为每个智能体设计细化的中间任务，逐步引导学习
- 在MPE和捉迷藏环境中，性能优于现有方法
- 适合研究稀疏奖励下的多智能体协作问题
稀疏奖励环境给强化学习带来巨大挑战，尤其在多智能体系统中，反馈延迟且共享，导致学习效果不佳。我们提出协同多维课程学习（CCL），一种新型课程学习框架，通过（1）优化个体智能体的中间任务，（2）使用变分进化算法生成有信息量的子任务，（3）协同演化智能体与环境以增强训练稳定性。在MPE和捉迷藏环境中的五个协作任务上实验表明，CCL在稀疏奖励设置下显著优于现有方法。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。