Spectral Model eXplainer: a chemically-grounded explainability framework for spectral-based machine learning models

TL;DR

The paper introduces SMX, a spectral explainability framework that leverages expert-informed spectral zones and PCA to provide chemically meaningful, visualizable explanations for spectral classifiers.

Contribution

SMX is a novel, model-agnostic, post-hoc XAI method that explains spectral models through spectral zones and predicate relevance, addressing limitations of existing tools.

Findings

SMX effectively explains spectral classifiers across multiple datasets.

The method provides direct visual comparison with measured spectra.

SMX outperforms traditional variable importance methods in spectral interpretability.

Abstract

Spectral-based machine learning models have been increasingly deployed in chemometrics and spectroscopy, where predictive accuracy is as important as explainability. Current employed eXplainable Artificial Intelligence (XAI) methods are largely adapted from tabular or generic multivariate domains, assigning relevance to isolated spectral variables rather than to the chemically meaningful spectral zones. Widely adopted tools such as SHapley Additive exPlanations (SHAP), Permutation Feature Importance (PFI), and Variable Importance in Projection scores (VIP) were not designed for the physical continuity and high collinearity of spectral data, and their variable-level outputs require post-hoc aggregation to recover zone-level information. This study introduces the Spectral Model eXplainer (SMX), a post-hoc, global, model-agnostic XAI framework that explains spectral classifiers through expert-informed spectral zones. SMX summarizes each zone via PCA, defines quantile-based logical predicates, estimates predicate relevance with perturbation in stochastic subsamples, and aggregates bag-wise rankings in a directed weighted graph summarized by Local Reaching Centrality. A key component is threshold spectrum reconstruction, which back-projects predicate boundaries to the original spectral domain in natural measurement units, enabling direct visual comparison with measured spectra. The method was evaluated on eight real spectral datasets (six based on X-ray Fluorescence--XRF and two based on Gamma-ray Spectrometry) and one synthetic benchmark with known gr