TY - GEN
T1 - Higher-Order Neural Additive Models
T2 - 25th IEEE International Conference on Data Mining, ICDM 2025
AU - Kim, Minkyu
AU - Choi, Hyun Soo
AU - Kim, Jinho
N1 - Publisher Copyright:
© 2025 IEEE.
PY - 2025
Y1 - 2025
N2 - 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/.
AB - 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/.
KW - Feature Interactions
KW - Generalized Additive Model
KW - Interpretability
KW - Interpretable Machine Learning
UR - https://www.scopus.com/pages/publications/105035065480
U2 - 10.1109/ICDM65498.2025.00140
DO - 10.1109/ICDM65498.2025.00140
M3 - Conference contribution
AN - SCOPUS:105035065480
T3 - Proceedings - IEEE International Conference on Data Mining, ICDM
SP - 1310
EP - 1319
BT - Proceedings - 25th IEEE International Conference on Data Mining, ICDM 2025
A2 - Ding, Wei
A2 - Vreeken, Jilles
A2 - Lu, Chang-Tien
A2 - Gunopulos, Dimitrios
A2 - Wu, Xindong
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 12 November 2025 through 15 November 2025
ER -