A Review of Uncertainty for Deep Reinforcement Learning

TL;DR

This paper reviews existing techniques for handling uncertainty in deep reinforcement learning, highlighting their importance, current state, and empirical benefits across various tasks to guide future research.

Contribution

It provides a comprehensive overview of uncertainty-aware methods in deep reinforcement learning, consolidating disparate results to foster further advancements.

Findings

Empirical benefits demonstrated across multiple RL tasks.

Uncertainty techniques improve robustness of RL agents.

The review promotes future research directions.

Abstract

Uncertainty is ubiquitous in games, both in the agents playing games and often in the games themselves. Working with uncertainty is therefore an important component of successful deep reinforcement learning agents. While there has been substantial effort and progress in understanding and working with uncertainty for supervised learning, the body of literature for uncertainty aware deep reinforcement learning is less developed. While many of the same problems regarding uncertainty in neural networks for supervised learning remain for reinforcement learning, there are additional sources of uncertainty due to the nature of an interactable environment. In this work, we provide an overview motivating and presenting existing techniques in uncertainty aware deep reinforcement learning. These works show empirical benefits on a variety of reinforcement learning tasks. This work serves to help to centralize the disparate results and promote future research in this area.