Enhancing RL Generalizability in Robotics through SHAP Analysis of Algorithms and Hyperparameters

TL;DR

This paper introduces an explainable SHAP-based framework to analyze and improve the generalization of reinforcement learning algorithms in robotics by quantifying configuration impacts.

Contribution

It provides a systematic, quantitative method to evaluate and select RL configurations for better generalization across robotic environments.

Findings

SHAP values reveal consistent configuration impact patterns across tasks.

Applying SHAP-guided selection improves RL generalization performance.

Theoretical link between Shapley values and RL generalizability established.

Abstract

Despite significant advances in Reinforcement Learning (RL), model performance remains highly sensitive to algorithm and hyperparameter configurations, while generalization gaps across environments complicate real-world deployment. Although prior work has studied RL generalization, the relative contribution of specific configurations to the generalization gap has not been quantitatively decomposed and systematically leveraged for configuration selection. To address this limitation, we propose an explainable framework that evaluates RL performance across robotic environments using SHapley Additive exPlanations (SHAP) to quantify configuration impacts. We establish a theoretical foundation connecting Shapley values to generalizability, empirically analyze configuration impact patterns, and introduce SHAP-guided configuration selection to enhance generalization. Our results reveal distinct patterns across algorithms and hyperparameters, with consistent configuration impacts across diverse tasks and environments. By applying these insights to configuration selection, we achieve improved RL generalizability and provide actionable guidance for practitioners.