ReCCoVER: Detecting Causal Confusion for Explainable Reinforcement Learning

TL;DR

ReCCoVER is a method for detecting when reinforcement learning agents rely on spurious correlations, improving transparency and guiding feature selection to prevent causal confusion in high-risk applications.

Contribution

The paper introduces ReCCoVER, an algorithm that detects causal confusion in DRL agents by testing their policies in alternative environments, enhancing explainability and safety.

Findings

ReCCoVER successfully identifies states with causal confusion.

The method provides actionable feature recommendations.

Demonstrated effectiveness in taxi and grid world environments.

Abstract

Despite notable results in various fields over the recent years, deep reinforcement learning (DRL) algorithms lack transparency, affecting user trust and hindering their deployment to high-risk tasks. Causal confusion refers to a phenomenon where an agent learns spurious correlations between features which might not hold across the entire state space, preventing safe deployment to real tasks where such correlations might be broken. In this work, we examine whether an agent relies on spurious correlations in critical states, and propose an alternative subset of features on which it should base its decisions instead, to make it less susceptible to causal confusion. Our goal is to increase transparency of DRL agents by exposing the influence of learned spurious correlations on its decisions, and offering advice to developers about feature selection in different parts of state space, to avoid causal confusion. We propose ReCCoVER, an algorithm which detects causal confusion in agent's reasoning before deployment, by executing its policy in alternative environments where certain correlations between features do not hold. We demonstrate our approach in taxi and grid world environments, where ReCCoVER detects states in which an agent relies on spurious correlations and offers a set of features that should be considered instead.