arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.23804cs.LG
收藏

行为克隆预训练用于近端策略优化

Actor-Critic Pretraining for Proximal Policy Optimization

Institute for Control Engineering of Machine Tools and Manufacturing Units ISW, University of Stuttgart(斯图加特大学机床与制造装置控制工程研究所ISW);Materials Testing Institute MPA, University of Stuttgart(斯图加特大学材料测试研究所MPA)

Andreas Kernbach, Amr Elsheikh, Nicolas Grupp, René Nagel, Marco F. Huber

AI总结:

本文提出了一种基于行为克隆的actor-critic预训练方法,用于近端策略优化,通过初始化actor和critic网络提升样本效率。

摘要:

强化学习(RL)的actor-critic算法能够实现自主学习,但通常需要大量的环境交互,这限制了其在机器人领域的应用。利用专家数据可以减少所需的环境交互次数。一种常见方法是actor预训练,其中actor网络通过行为克隆在专家演示上进行初始化,随后通过RL进行微调。相比之下,critic网络的初始化受到较少关注,尽管其在策略优化中起着核心作用。本文提出了一种用于actor-critic算法(如近端策略优化(PPO))的预训练方法,该方法利用专家演示来初始化两个网络。actor通过行为克隆进行预训练,而critic则通过预训练策略的回放获得的回报进行预训练。该方法在15个模拟的机器人操作和运动任务上进行了评估。实验结果表明,actor-critic预训练在平均情况下比无预训练提高了86.1%的样本效率,比仅actor预训练提高了30.9%。

英文摘要:

Reinforcement learning (RL) actor-critic algorithms enable autonomous learning but often require a large number of environment interactions, which limits their applicability in robotics. Leveraging expert data can reduce the number of required environment interactions. A common approach is actor pretraining, where the actor network is initialized via behavioral cloning on expert demonstrations and subsequently fine-tuned with RL. In contrast, the initialization of the critic network has received little attention, despite its central role in policy optimization. This paper proposes a pretraining approach for actor-critic algorithms like Proximal Policy Optimization (PPO) that uses expert demonstrations to initialize both networks. The actor is pretrained via behavioral cloning, while the critic is pretrained using returns obtained from rollouts of the pretrained policy. The approach is evaluated on 15 simulated robotic manipulation and locomotion tasks. Experimental results show that actor-critic pretraining improves sample efficiency by 86.1% on average compared to no pretraining and by 30.9% to actor-only pretraining.

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。

↑