arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.07358cs.LGstat.ML
收藏

学习在经验回放缓冲区中使用本地和全局上下文进行采样

Learning to Sample with Local and Global Contexts in Experience Replay Buffer

Samsung Advanced Institute of Technology(三星综合技术院);University of California, Berkeley(加州大学伯克利分校);Korea Advanced Institute of Science and Technology(韩国科学技术院);AITRICS

Youngmin Oh, Kimin Lee, Jinwoo Shin, Eunho Yang, Sung Ju Hwang

摘要:

经验重放使智能体能够记住和重用过去的经验,在非策略强化学习(RL)的成功中发挥了重要作用。为了有效地利用经验重放,现有的采样方法允许通过基于某些度量(例如TD误差)对它们施加优先级来选择出更有意义的经验。然而,它们可能导致采样高度偏置的冗余转换,因为它们独立地计算每个转换的采样率,而不考虑其相对于其他转换的重要性。在本文中,我们的目标是解决这个问题,提出了一个新的基于学习的抽样方法,可以计算过渡的相对重要性。为此,我们设计了一种新的置换等变神经架构,不仅需要从每个过渡(本地)的功能,但也作为输入的其他(全球)的上下文。我们验证了我们的框架,我们称之为神经经验重放采样器(NERS),在多个基准任务的连续和离散控制任务,并表明它可以显着提高性能的各种非政策RL方法。进一步的分析证实,样本效率的提高确实是由于采样的多样性和有意义的过渡NERS考虑本地和全球的背景。

英文摘要:

Experience replay, which enables the agents to remember and reuse experience from the past, has played a significant role in the success of off-policy reinforcement learning (RL). To utilize the experience replay efficiently, the existing sampling methods allow selecting out more meaningful experiences by imposing priorities on them based on certain metrics (e.g. TD-error). However, they may result in sampling highly biased, redundant transitions since they compute the sampling rate for each transition independently, without consideration of its importance in relation to other transitions. In this paper, we aim to address the issue by proposing a new learning-based sampling method that can compute the relative importance of transition. To this end, we design a novel permutation-equivariant neural architecture that takes contexts from not only features of each transition (local) but also those of others (global) as inputs. We validate our framework, which we refer to as Neural Experience Replay Sampler (NERS), on multiple benchmark tasks for both continuous and discrete control tasks and show that it can significantly improve the performance of various off-policy RL methods. Further analysis confirms that the improvements of the sample efficiency indeed are due to sampling diverse and meaningful transitions by NERS that considers both local and global contexts.

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。

↑