Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data AugmentationDownload PDF

Published: 22 Jul 2021, Last Modified: 22 Oct 2023URL 2021 PosterReaders: Everyone
Keywords: reinforcement learning, generalization, data augmentation
TL;DR: In this work, we illuminate causes of instability in deep Q-learning from augmented data, and propose a simple yet effective method that improves stability, sample efficiency, and generalization over prior methods.
Abstract: While agents trained by Reinforcement Learning (RL) can solve increasingly challenging tasks directly from visual observations, generalizing learned skills to novel environments remains very challenging. Extensive use of data augmentation is a promising technique for improving generalization in RL, but it is often found to decrease sample efficiency and can even lead to divergence. In this paper, we investigate causes of instability when using data augmentation in common off-policy RL algorithms. We identify two problems, both rooted in high-variance Q-targets. Based on our findings, we propose a simple yet effective technique for stabilizing this class of algorithms under augmentation. We perform extensive empirical evaluation of image-based RL using both ConvNets and Vision Transformers (ViT) on a family of benchmarks based on DeepMind Control Suite, as well as in robotic manipulation tasks. Our method greatly improves stability and sample efficiency of ConvNets under augmentation, and achieves generalization results competitive with state-of-the-art methods for image-based RL. We further show that our method scales to RL with ViT-based architectures, and that data augmentation may be especially important in this setting. Code and videos: https://nicklashansen.github.io/SVEA
Community Implementations: [![CatalyzeX](/images/catalyzex_icon.svg) 1 code implementation](https://www.catalyzex.com/paper/arxiv:2107.00644/code)
1 Reply

Loading