English

Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels

Machine Learning 2021-03-09 v4 Computer Vision and Pattern Recognition Image and Video Processing Machine Learning

Abstract

We propose a simple data augmentation technique that can be applied to standard model-free reinforcement learning algorithms, enabling robust learning directly from pixels without the need for auxiliary losses or pre-training. The approach leverages input perturbations commonly used in computer vision tasks to regularize the value function. Existing model-free approaches, such as Soft Actor-Critic (SAC), are not able to train deep networks effectively from image pixels. However, the addition of our augmentation method dramatically improves SAC's performance, enabling it to reach state-of-the-art performance on the DeepMind control suite, surpassing model-based (Dreamer, PlaNet, and SLAC) methods and recently proposed contrastive learning (CURL). Our approach can be combined with any model-free reinforcement learning algorithm, requiring only minor modifications. An implementation can be found at https://sites.google.com/view/data-regularized-q.

Keywords

Cite

@article{arxiv.2004.13649,
  title  = {Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels},
  author = {Ilya Kostrikov and Denis Yarats and Rob Fergus},
  journal= {arXiv preprint arXiv:2004.13649},
  year   = {2021}
}
R2 v1 2026-06-23T15:09:31.831Z