Neural-ANOVA: Model Decomposition for Interpretable Machine Learning

The analysis of variance (ANOVA) decomposition offers a systematic method to understand the interaction effects that contribute to a specific decision output. In this paper we introduce Neural-ANOVA, an approach to decompose neural networks into glassbox models using the ANOVA decomposition. Our app...

Full description

Saved in:
Bibliographic Details
Main Authors Limmer, Steffen, Udluft, Steffen, Otte, Clemens
Format Journal Article
LanguageEnglish
Published 22.08.2024
Subjects
Computer Science - Learning
Statistics - Machine Learning
Online AccessGet full text

Cover

More Information
Summary:The analysis of variance (ANOVA) decomposition offers a systematic method to understand the interaction effects that contribute to a specific decision output. In this paper we introduce Neural-ANOVA, an approach to decompose neural networks into glassbox models using the ANOVA decomposition. Our approach formulates a learning problem, which enables rapid and closed-form evaluation of integrals over subspaces that appear in the calculation of the ANOVA decomposition. Finally, we conduct numerical experiments to illustrate the advantages of enhanced interpretability and model validation by a decomposition of the learned interaction effects.
DOI:10.48550/arxiv.2408.12319