DeepAI AI Chat
Log In Sign Up

Robust Deep Reinforcement Learning via Multi-View Information Bottleneck

by   Jiameng Fan, et al.

Deep reinforcement learning (DRL) agents are often sensitive to visual changes that were unseen in their training environments. To address this problem, we introduce a robust representation learning approach for RL. We introduce an auxiliary objective based on the multi-view information bottleneck (MIB) principle which encourages learning representations that are both predictive of the future and less sensitive to task-irrelevant distractions. This enables us to train high-performance policies that are robust to visual distractions and can generalize to unseen environments. We demonstrate that our approach can achieve SOTA performance on challenging visual control tasks, even when the background is replaced with natural videos. In addition, we show that our approach outperforms well-established baselines on generalization to unseen environments using the large-scale Procgen benchmark.


page 6

page 7

page 16


Dynamics Generalization via Information Bottleneck in Deep Reinforcement Learning

Despite the significant progress of deep reinforcement learning (RL) in ...

Control of Memory, Active Perception, and Action in Minecraft

In this paper, we introduce a new set of reinforcement learning (RL) tas...

Improving Generalization in Reinforcement Learning with Mixture Regularization

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

Temporal Disentanglement of Representations for Improved Generalisation in Reinforcement Learning

In real-world robotics applications, Reinforcement Learning (RL) agents ...

An Actor-Critic-Attention Mechanism for Deep Reinforcement Learning in Multi-view Environments

In reinforcement learning algorithms, leveraging multiple views of the e...

Accelerating Representation Learning with View-Consistent Dynamics in Data-Efficient Reinforcement Learning

Learning informative representations from image-based observations is of...

Learning Task-Driven Control Policies via Information Bottlenecks

This paper presents a reinforcement learning approach to synthesizing ta...