Introspection-based Explainable Reinforcement Learning in Episodic and Non-episodic Scenarios

TL;DR

This paper develops and evaluates an introspection-based explainable reinforcement learning approach for robotic tasks, enhancing understanding and trust in both episodic and non-episodic scenarios through probability-based explanations.

Contribution

It introduces a normalization step for Q-values, enabling introspection-based explanations in environments with negative or small Q-values, applicable to both episodic and non-episodic robotics tasks.

Findings

Effective in episodic robotics simulations

Applicable to non-episodic environments

Normalization improves explanation quality

Abstract

With the increasing presence of robotic systems and human-robot environments in today's society, understanding the reasoning behind actions taken by a robot is becoming more important. To increase this understanding, users are provided with explanations as to why a specific action was taken. Among other effects, these explanations improve the trust of users in their robotic partners. One option for creating these explanations is an introspection-based approach which can be used in conjunction with reinforcement learning agents to provide probabilities of success. These can in turn be used to reason about the actions taken by the agent in a human-understandable fashion. In this work, this introspection-based approach is developed and evaluated further on the basis of an episodic and a non-episodic robotics simulation task. Furthermore, an additional normalization step to the Q-values is proposed, which enables the usage of the introspection-based approach on negative and comparatively small Q-values. Results obtained show the viability of introspection for episodic robotics tasks and, additionally, that the introspection-based approach can be used to generate explanations for the actions taken in a non-episodic robotics environment as well.