Minimizing False-Positive Attributions in Explanations of Non-Linear Models

TL;DR

PatternLocal is a new XAI method that reduces false-positive attributions caused by suppressor variables in non-linear models, improving explanation reliability and interpretability.

Contribution

It introduces PatternLocal, a technique that transforms local surrogate models into a generative form to suppress suppressor variables in non-linear explanations.

Findings

Outperformed existing XAI methods on the XAI-TRIS benchmark.

Reduced false-positive attributions in non-linear model explanations.

Produced physiologically plausible explanations on EEG data.

Abstract

Suppressor variables can influence model predictions without being dependent on the target outcome, and they pose a significant challenge for Explainable AI (XAI) methods. These variables may cause false-positive feature attributions, undermining the utility of explanations. Although effective remedies exist for linear models, their extension to non-linear models and instance-based explanations has remained limited. We introduce PatternLocal, a novel XAI technique that addresses this gap. PatternLocal begins with a locally linear surrogate, e.g., LIME, KernelSHAP, or gradient-based methods, and transforms the resulting discriminative model weights into a generative representation, thereby suppressing the influence of suppressor variables while preserving local fidelity. In extensive hyperparameter optimization on the XAI-TRIS benchmark, PatternLocal consistently outperformed other XAI methods and reduced false-positive attributions when explaining non-linear tasks, thereby enabling more reliable and actionable insights. We further evaluate PatternLocal on an EEG motor imagery dataset, demonstrating physiologically plausible explanations.