analiza aiqbench
This paper introduces HGNAN, a framework that applies neural additive models to hypergraphs to achieve both high predictive performance and intrinsic interpretability in higher-order relational data. By decomposing feature-wise nonlinearities alongside structural aggregation, it addresses the black-box nature of current hypergraph neural networks without sacrificing accuracy on benchmark datasets. This work advances the field of machine learning by offering transparent methods for analyzing complex network dependencies where traditional message-passing architectures fall short.
Interpretable Hypergraph Learning via Neural Additive Models
arXiv:2610.07458v1 Announce Type: new Abstract: Hypergraphs offer a natural framework for modeling networked data, where dependencies among entities are governed by higher-order interactions. While hypergraph learning methods such as hypergraph neural networks have demonstrated remarkable predictive performance, most existing approaches rely on black-box message-passing architectures, making it difficult to disentangle the contributions of node attributes and higher-order structural information. To address this challenge, we introduce the hypergraph neural additive network (HGNAN), an inherently interpretable framework for learning on hypergraph-structured data. HGNAN extends classical neural additive models to higher-order relational data by integrating feature-wise nonlinear decomposition with hypergraph-aware structural aggregation, enabling transparent prediction for both node- and hyperedge-level tasks. Extensive experiments on benchmark datasets demonstrate that HGNAN achieves performance comparable with state-of-the-art hypergraph learning methods while providing intrinsic and meaningful interpretability.
Abstract:Hypergraphs offer a natural framework for modeling networked data, where dependencies among entities are governed by higher-order interactions. While hypergraph learning methods such as hypergraph neural networks have demonstrated remarkable predictive performance, most existing approaches rely on black-box message-passing architectures, making it difficult to disentangle the contributions of node attributes and higher-order structural information. To address this challenge, we introduce the hypergraph neural additive network (HGNAN), an inherently interpretable framework for learning on hypergraph-structured data. HGNAN extends classical neural additive models to higher-order relational data by integrating feature-wise nonlinear decomposition with hypergraph-aware structural aggregation, enabling transparent prediction for both node- and hyperedge-level tasks. Extensive experiments on benchmark datasets demonstrate that HGNAN achieves performance comparable with state-of-the-art hypergraph learning methods while providing intrinsic and meaningful interpretability.
Machine Learning (cs.LG); Social and Information Networks (cs.SI)
arXiv:2610.07458 [cs.LG]
(or arXiv:2610.07458v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.07458 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 224 cuvinte
Subiecte conexe
Generalization and infinite width
LessWrong · 1 aug. 2026
PIRAMID: Progress and Plans
LessWrong · 27 iul. 2026
Interpretable and Fair Generalized Additive Neural Networks via Multi-objective Learning
arXiv · 9 sept. 2026
Latent reasoning architectures would likely undermine CoT, our strongest oversight tool
AI Alignment Forum · 23 sept. 2026