arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.06386cs.LGcs.AI
收藏

用自监督轨迹对比学习改进基于上下文的Meta-强化学习

Improving Context-Based Meta-Reinforcement Learning with Self-Supervised Trajectory Contrastive Learning

University of California, Berkeley(加利福尼亚大学伯克利分校);Tsinghua University(清华大学);Facebook Research(Facebook研究院)

Bernie Wang, Simon Xu, Kurt Keutzer, Yang Gao, Bichen Wu

摘要:

Meta-强化学习通常需要比单任务强化学习方法多几个数量级的样本。这是因为元训练需要处理更多样化的分布和训练额外的组件,如上下文编码器。为了解决这个问题,我们提出了一种新的自监督学习任务,我们称之为轨迹对比学习(TCL),以改善元训练。TCL采用对比学习并训练上下文编码器来预测是否从同一轨迹采样两个过渡窗口。TCL利用了基于上下文的元RL的自然层次结构,并做出了最少的假设,使其能够普遍适用于基于上下文的元RL算法。它加速了上下文编码器的训练,并从整体上改进了元训练。实验表明,在大多数环境中,TCL在元RL MuJoCo(6个中的5个)和元世界基准测试(50个中的44个)上的表现都优于或优于强元RL基线。

英文摘要:

Meta-reinforcement learning typically requires orders of magnitude more samples than single task reinforcement learning methods. This is because meta-training needs to deal with more diverse distributions and train extra components such as context encoders. To address this, we propose a novel self-supervised learning task, which we named Trajectory Contrastive Learning (TCL), to improve meta-training. TCL adopts contrastive learning and trains a context encoder to predict whether two transition windows are sampled from the same trajectory. TCL leverages the natural hierarchical structure of context-based meta-RL and makes minimal assumptions, allowing it to be generally applicable to context-based meta-RL algorithms. It accelerates the training of context encoders and improves meta-training overall. Experiments show that TCL performs better or comparably than a strong meta-RL baseline in most of the environments on both meta-RL MuJoCo (5 of 6) and Meta-World benchmarks (44 out of 50).

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。

↑