Interpretable pipelines with evolutionarily optimized modules for RL tasks with visual inputs

TL;DR

This paper introduces evolutionarily optimized, interpretable pipelines for reinforcement learning with visual inputs, decomposing decision processes into feature extraction and reasoning, achieving comparable performance to black-box models in Atari environments.

Contribution

The paper presents a novel end-to-end pipeline of interpretable models optimized by evolutionary algorithms for RL tasks with raw visual data.

Findings

Comparable performance to black-box models without stochastic frame-skipping

Performance degradation observed with frame-skipping

Effective decomposition of decision-making into feature extraction and reasoning

Abstract

The importance of explainability in AI has become a pressing concern, for which several explainable AI (XAI) approaches have been recently proposed. However, most of the available XAI techniques are post-hoc methods, which however may be only partially reliable, as they do not reflect exactly the state of the original models. Thus, a more direct way for achieving XAI is through interpretable (also called glass-box) models. These models have been shown to obtain comparable (and, in some cases, better) performance with respect to black-boxes models in various tasks such as classification and reinforcement learning. However, they struggle when working with raw data, especially when the input dimensionality increases and the raw inputs alone do not give valuable insights on the decision-making process. Here, we propose to use end-to-end pipelines composed of multiple interpretable models co-optimized by means of evolutionary algorithms, that allows us to decompose the decision-making process into two parts: computing high-level features from raw data, and reasoning on the extracted high-level features. We test our approach in reinforcement learning environments from the Atari benchmark, where we obtain comparable results (with respect to black-box approaches) in settings without stochastic frame-skipping, while performance degrades in frame-skipping settings.