Skip to main navigation Skip to search Skip to main content

Higher-Order Neural Additive Models: An Interpretable Machine Learning Model with Feature Interactions

  • Ltd.
  • Kangwon National University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Neural Additive Models (NAMs) have recently demonstrated promising predictive performance while maintaining interpretability. However, their capacity is limited to capturing only first-order feature interactions, which restricts their effectiveness on real-world datasets. To address this limitation, we propose Higher-order Neural Additive Models (HONAMs), an interpretable machine learning model that effectively and efficiently captures feature interactions of arbitrary orders. HONAMs improve predictive accuracy without compromising interpretability, an essential requirement in high-stakes applications. This advantage of HONAM can help analyze and extract high-order interactions present in datasets. The source code for HONAM is publicly available at https://github.com/gim4855744/HONAM/.

Original languageEnglish
Title of host publicationProceedings - 25th IEEE International Conference on Data Mining, ICDM 2025
EditorsWei Ding, Jilles Vreeken, Chang-Tien Lu, Dimitrios Gunopulos, Xindong Wu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1310-1319
Number of pages10
ISBN (Electronic)9798331595999
DOIs
StatePublished - 2025
Event25th IEEE International Conference on Data Mining, ICDM 2025 - Washington, United States
Duration: 12 Nov 202515 Nov 2025

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
ISSN (Print)1550-4786

Conference

Conference25th IEEE International Conference on Data Mining, ICDM 2025
Country/TerritoryUnited States
CityWashington
Period12/11/2515/11/25

Keywords

  • Feature Interactions
  • Generalized Additive Model
  • Interpretability
  • Interpretable Machine Learning

Fingerprint

Dive into the research topics of 'Higher-Order Neural Additive Models: An Interpretable Machine Learning Model with Feature Interactions'. Together they form a unique fingerprint.

Cite this