arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01196cs.LG
收藏

揭示递归神经策略中的隐藏动力学结构

Unraveling the Hidden Dynamical Structure in Recurrent Neural Policies

University of Science and Technology of China(中国科学技术大学)

Jin Li, Yue Wu, Mengsha Huang, Yuhao Sun, Hao He, Xianyuan Zhan

AI总结:

本文通过分析不同训练方法、模型架构和任务中学习得到的递归策略的隐藏状态域,发现稳定的循环结构在与环境交互时 consistently 出现,这些结构与动力系统分析中的极限环有相似性,并揭示了极限环几何结构与策略行为之间的对应关系,为解释递归策略的性能提供了新视角。

摘要:

递归神经策略在部分可观测控制和元强化学习任务中被广泛应用。它们能够维持内部记忆并快速适应未见过的场景,相较于非递归策略具有无可比拟的性能。然而,到目前为止,其优异的泛化性和鲁棒性性能的底层机制仍不明确。在本研究中,通过分析不同训练方法、模型架构和任务中学习得到的递归策略的隐藏状态域,我们发现稳定的循环结构在与环境交互时 consistently 出现。如果将策略和环境视为一个联合混合动力系统,这些循环结构与动力系统分析中的极限环有显著相似性。此外,我们发现这些极限环的几何结构也与策略行为具有结构化的对应关系。这些发现为解释递归策略的许多良好特性提供了新的视角:极限环的出现稳定了策略的内部记忆和任务相关的环境状态,同时抑制了来自环境不确定性的干扰变量;极限环的几何结构也编码了行为的关联结构,有助于在非稳态环境中更轻松地进行技能适应。

英文摘要:

Recurrent neural policies are widely used in partially observable control and meta-RL tasks. Their abilities to maintain internal memory and adapt quickly to unseen scenarios have offered them unparalleled performance when compared to non-recurrent counterparts. However, until today, the underlying mechanisms for their superior generalization and robustness performance remain poorly understood. In this study, by analyzing the hidden state domain of recurrent policies learned over a diverse set of training methods, model architectures, and tasks, we find that stable cyclic structures consistently emerge during interaction with the environment. Such cyclic structures share a remarkable similarity with \textit{limit cycles} in dynamical system analysis, if we consider the policy and the environment as a joint hybrid dynamical system. Moreover, we uncover that the geometry of such limit cycles also has a structured correspondence with the policies' behaviors. These findings offer new perspectives to explain many nice properties of recurrent policies: the emergence of limit cycles stabilizes both the policies' internal memory and the task-relevant environmental states, while suppressing nuisance variability arising from environmental uncertainty; the geometry of limit cycles also encodes relational structures of behaviors, facilitating easier skill adaptation when facing non-stationary environments.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑