A Survey of Explainable Reinforcement Learning

TL;DR

This survey reviews explainable reinforcement learning (XRL), proposing a taxonomy, summarizing current techniques, identifying gaps, and outlining future research directions in the field.

Contribution

It introduces a novel taxonomy for XRL, organizes existing techniques accordingly, and highlights research gaps and future directions.

Findings

Proposes a taxonomy prioritizing RL settings

Summarizes current XRL techniques

Identifies gaps and future research directions

Abstract

Explainable reinforcement learning (XRL) is an emerging subfield of explainable machine learning that has attracted considerable attention in recent years. The goal of XRL is to elucidate the decision-making process of learning agents in sequential decision-making settings. In this survey, we propose a novel taxonomy for organizing the XRL literature that prioritizes the RL setting. We overview techniques according to this taxonomy. We point out gaps in the literature, which we use to motivate and outline a roadmap for future work.