Towards Piece-by-Piece Explanations for Chess Positions with SHAP

TL;DR

This paper adapts SHAP, an explainable AI technique, to attribute chess engine evaluations to individual pieces, enhancing interpretability and aiding human understanding of engine decisions.

Contribution

It introduces a novel method applying SHAP to chess, providing locally faithful, human-interpretable explanations of engine evaluations based on piece contributions.

Findings

Enables visualization of piece contributions in chess positions

Improves human understanding of engine evaluations

Provides a foundation for interpretable chess AI research

Abstract

Contemporary chess engines offer precise yet opaque evaluations, typically expressed as centipawn scores. While effective for decision-making, these outputs obscure the underlying contributions of individual pieces or patterns. In this paper, we explore adapting SHAP (SHapley Additive exPlanations) to the domain of chess analysis, aiming to attribute a chess engines evaluation to specific pieces on the board. By treating pieces as features and systematically ablating them, we compute additive, per-piece contributions that explain the engines output in a locally faithful and human-interpretable manner. This method draws inspiration from classical chess pedagogy, where players assess positions by mentally removing pieces, and grounds it in modern explainable AI techniques. Our approach opens new possibilities for visualization, human training, and engine comparison. We release accompanying code and data to foster future research in interpretable chess AI.