Decoupling Representation Learning from Reinforcement Learning

by   Adam Stooke, et al.
berkeley college

In an effort to overcome limitations of reward-driven feature learning in deep reinforcement learning (RL) from images, we propose decoupling representation learning from policy learning. To this end, we introduce a new unsupervised learning (UL) task, called Augmented Temporal Contrast (ATC), which trains a convolutional encoder to associate pairs of observations separated by a short time difference, under image augmentations and using a contrastive loss. In online RL experiments, we show that training the encoder exclusively using ATC matches or outperforms end-to-end RL in most environments. Additionally, we benchmark several leading UL algorithms by pre-training encoders on expert demonstrations and using them, with weights frozen, in RL agents; we find that agents using ATC-trained encoders outperform all others. We also train multi-task encoders on data from multiple environments and show generalization to different downstream RL tasks. Finally, we ablate components of ATC, and introduce a new data augmentation to enable replay of (compressed) latent images from pre-trained encoders when RL requires augmentation. Our experiments span visually diverse RL benchmarks in DeepMind Control, DeepMind Lab, and Atari, and our complete code is available at


page 8

page 13


Cross-domain Random Pre-training with Prototypes for Reinforcement Learning

Task-agnostic cross-domain pre-training shows great potential in image-b...

Task-Induced Representation Learning

In this work, we evaluate the effectiveness of representation learning a...

CRC-RL: A Novel Visual Feature Representation Architecture for Unsupervised Reinforcement Learning

This paper addresses the problem of visual feature representation learni...

Reinforcement Learning with Augmented Data

Learning from visual observations is a fundamental yet challenging probl...

Deep Reinforcement Learning for Cost-Effective Medical Diagnosis

Dynamic diagnosis is desirable when medical tests are costly or time-con...

Improving Generalization in Reinforcement Learning with Mixture Regularization

Deep reinforcement learning (RL) agents trained in a limited set of envi...

Stabilizing Off-Policy Deep Reinforcement Learning from Pixels

Off-policy reinforcement learning (RL) from pixel observations is notori...

Please sign up or login with your details

Forgot password? Click here to reset