Salvato in:
Dettagli Bibliografici
Autori principali: Fan, Wangxuan, Wang, Ching, Li, Siqi, Liu, Nan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.01663
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916985365856256
author Fan, Wangxuan
Wang, Ching
Li, Siqi
Liu, Nan
author_facet Fan, Wangxuan
Wang, Ching
Li, Siqi
Liu, Nan
contents For many real-world applications, understanding feature-outcome relationships is as crucial as achieving high predictive accuracy. While traditional neural networks excel at prediction, their black-box nature obscures underlying functional relationships. Kolmogorov--Arnold Networks (KANs) address this by employing learnable spline-based activation functions on edges, enabling recovery of symbolic representations while maintaining competitive performance. However, KAN's architecture presents unique challenges for network pruning. Conventional magnitude-based methods become unreliable due to sensitivity to input coordinate shifts. We propose \textbf{ShapKAN}, a pruning framework using Shapley value attribution to assess node importance in a shift-invariant manner. Unlike magnitude-based approaches, ShapKAN quantifies each node's actual contribution, ensuring consistent importance rankings regardless of input parameterization. Extensive experiments on synthetic and real-world datasets demonstrate that ShapKAN preserves true node importance while enabling effective network compression. Our approach improves KAN's interpretability advantages, facilitating deployment in resource-constrained environments.
format Preprint
id arxiv_https___arxiv_org_abs_2510_01663
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Shift-Invariant Attribute Scoring for Kolmogorov-Arnold Networks via Shapley Value
Fan, Wangxuan
Wang, Ching
Li, Siqi
Liu, Nan
Machine Learning
Artificial Intelligence
For many real-world applications, understanding feature-outcome relationships is as crucial as achieving high predictive accuracy. While traditional neural networks excel at prediction, their black-box nature obscures underlying functional relationships. Kolmogorov--Arnold Networks (KANs) address this by employing learnable spline-based activation functions on edges, enabling recovery of symbolic representations while maintaining competitive performance. However, KAN's architecture presents unique challenges for network pruning. Conventional magnitude-based methods become unreliable due to sensitivity to input coordinate shifts. We propose \textbf{ShapKAN}, a pruning framework using Shapley value attribution to assess node importance in a shift-invariant manner. Unlike magnitude-based approaches, ShapKAN quantifies each node's actual contribution, ensuring consistent importance rankings regardless of input parameterization. Extensive experiments on synthetic and real-world datasets demonstrate that ShapKAN preserves true node importance while enabling effective network compression. Our approach improves KAN's interpretability advantages, facilitating deployment in resource-constrained environments.
title Shift-Invariant Attribute Scoring for Kolmogorov-Arnold Networks via Shapley Value
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.01663