What is Your Force Field Really Learning? Gaining Scientific Intuition with A Dual-Level Explainability Framework

Yi Cao, Peter Mastracco, Jieneng Chen, Alan Yuille, Paulette Clancy

Track: Regular Track (Page limit: 6-8 pages)

Keywords: Machine Learning Force Fields (MLFFs), Explainable Artificial Intelligence (XAI), Model Interpretability, Scientific Machine Learning, Grad-CAM, SHAP, Material Science

TL;DR: DUAL-X, a dual-level explainability framework using Grad-CAM and SHAP, uncovers where and what ML force fields learn, enabling interpretable and generalizable physical reasoning in materials science.

Abstract:

Machine learning force fields (MLFFs) now rival quantum methods in accuracy, yet their internal logic remains opaque—a critical barrier to both trust and scientific discovery. We introduce **DUAL-X**, a *Dual-Level Explainability Framework* that bridges model reasoning with human understanding. **DUAL-X** unites two complementary perspectives: a **model-centric** level identifying *where* in the atomic structure the model focuses its attention, and a **human-centric** level revealing *what* physically meaningful interactions it prioritizes. Implemented with Grad-CAM for spatial attribution and SHAP-on-SOAP for physical interpretation, **DUAL-X** provides a general, human-in-the-loop paradigm for interpretable scientific AI.

Applied to dopant migration—a challenging task for MLFFs—**DUAL-X** reveals that training strategy governs the emergence of chemical intuition. Multi-temperature fine-tuned models exhibit over $10^2\times$ stronger selectivity for Cr–Cr $f$-type ($l=3$) angular correlations than scratch-trained ones, emphasizing complex 3D coordination motifs essential for accuracy. Models with sharper Grad-CAM focus also display coherent SHAP importance for dopant clustering, revealing consistent internal reasoning across scales.

By aligning model logic with human physical knowledge, **DUAL-X** transforms opaque predictors into interpretable scientific partners—advancing trustworthy, explainable, and insight-driven AI for materials discovery.

Submission Number: 12