Work

Interpreting Logical Explanations of Classifying Neural Networks

Year: 2026

Type: conference-paper

Abstract: Formal methods are routinely used to address the issue of explainability of machine learning models.Yet, it is not always trivial to understand how a logical explanation could be useful in practice due to human readability challenges.This paper applies classical geometric methods for interpreting logical explanations and illustrates the usefulness for the users on datasets from medical and image classification domains previously studied in the context of formal explainability. BackgroundThe computation of logical explanations of NN classification relies on logical solvers that yield a logical formula in the form of a sufficient condition that 375

Source: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN)

DOI: 10.14428/esann/2026.es2026-326

Authors: Fabrizio Leopardi, Faezeh Labbaf, Tomas Kolarik, Michael Wand, Natasha Sharygina

Institutions: Università della Svizzera italiana, Dalle Molle Institute for Artificial Intelligence Research

Language: English

Topic: Explainable Artificial Intelligence (XAI)

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: bronze

Funder: Hasler Stiftung

OpenAlex ID: W7154629944

FWCI: 8.936

Cites:

Cited by: 1

Related to: 10