A Short Survey on Probabilistic Reinforcement Learning

TL;DR

This paper surveys methods in reinforcement learning that balance exploration and exploitation, focusing on approaches that provide performance guarantees from fixed data samples, especially relevant in sensitive domains.

Contribution

It provides a concise overview of existing techniques for robust policy computation with fixed samples, highlighting approaches that ensure performance guarantees.

Findings

Summarizes key methods for exploration-exploitation trade-offs

Highlights techniques for robust policy computation from fixed data

Discusses applications in sensitive domains

Abstract

A reinforcement learning agent tries to maximize its cumulative payoff by interacting in an unknown environment. It is important for the agent to explore suboptimal actions as well as to pick actions with highest known rewards. Yet, in sensitive domains, collecting more data with exploration is not always possible, but it is important to find a policy with a certain performance guaranty. In this paper, we present a brief survey of methods available in the literature for balancing exploration-exploitation trade off and computing robust solutions from fixed samples in reinforcement learning.