arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.08588cs.LGcs.AI
收藏

面向连续控制的批量到流式深度强化学习

Towards Batch-to-Streaming Deep Reinforcement Learning for Continuous Control

Department of Information Engineering University of Padova(信息工程系帕多瓦大学)

Riccardo De Monte, Matteo Cederle, Gian Antonio Susto

AI总结:

本文提出两种流式深度强化学习算法,旨在与现有批量RL方法兼容,适用于设备端微调任务。算法在标准基准上表现优异,无需复杂超参数调优,并探讨了批量到流式转换的挑战。

摘要:

最先进的深度强化学习(RL)方法在连续控制任务中表现优异,但其计算复杂度与资源受限硬件不兼容,因依赖回放缓冲区、批量更新和目标网络。新兴的流式深度RL通过纯在线更新解决此问题,在标准基准上表现强劲。本文提出两种新算法:流式软演员-评论家(S2AC)和流式确定性演员-评论家(SDAC),专门设计与现有批量RL方法兼容,适用于设备端微调应用如Sim2Real迁移。两种算法在标准基准上表现与现有流式基线相当,无需繁琐的环境特定超参数调优。进一步研究了批量到流式转换,显示简单转换不保证预训练策略性能的保持,并提出解决此挑战的系统方法。

英文摘要:

State-of-the-art deep reinforcement learning (RL) methods have achieved remarkable performance in continuous control tasks, yet their computational complexity is often incompatible with the constraints of resource-limited hardware, due to their reliance on replay buffers, batch updates, and target networks. The emerging paradigm of streaming deep RL addresses this limitation through purely online updates, achieving strong empirical performance on standard benchmarks. In this work, we propose two novel streaming deep RL algorithms, Streaming Soft Actor-Critic (S2AC) and Streaming Deterministic Actor-Critic (SDAC), explicitly designed to be compatible with state-of-the-art batch RL methods, making them particularly suitable for on-device finetuning applications such as Sim2Real transfer. Both algorithms achieve performance comparable to state-of-the-art streaming baselines on standard benchmarks without requiring tedious per-environment hyperparameter tuning. We further investigate the batch-to-streaming transition, showing that a naive transition does not guarantee preservation of pre-trained policy performance, and propose a principled approach to address this challenge.

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。

↑