改进带记忆的策略梯度方法，提升部分可观测环境下的决策能力
Scaling Internal-State Policy-Gradient Methods for POMDPs
- 基于已知模型或仿真，直接优化带记忆的策略
- 在噪声机器人导航等大规模问题上显著优于传统方法
- 适合需要长期记忆的强化学习任务，如多智能体系统
策略梯度方法近年来受到广泛关注，用于在部分可观测环境中学习行动策略。尽管在无需记忆的场景中表现良好，但在需要记忆的场景中效果有限。本文提出几种改进算法，用于在无限时域设定下学习带记忆的策略——当环境模型已知时可直接使用，否则通过仿真实现。我们在多个大规模部分可观测马尔可夫决策过程（POMDP）问题上评估这些算法，包括噪声机器人导航和多智能体任务，结果表明新方法具有明显优势。
原文摘要 · Abstract (English)
Policy-gradient methods have received increased attention recently as a mechanism for learning to act in partially observable environments. They have shown promise for problems admitting memoryless policies but have been less successful when memory is required. In this paper we develop several improved algorithms for learning policies with memory in an infinite-horizon setting -- directly when a known model of the environment is available, and via simulation otherwise. We compare these algorithms on some large POMDPs, including noisy robot navigation and multi-agent problems.
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。