Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Bechler-Speicher, Maya, Zerio, Andrea, Huri, Maor, Vestergaard, Marie Vibeke, Gilad-Bachrach, Ran, Jess, Tine, Bhatt, Samir, Sazonovs, Aleksejs
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2509.23923
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917048809947136
author Bechler-Speicher, Maya
Zerio, Andrea
Huri, Maor
Vestergaard, Marie Vibeke
Gilad-Bachrach, Ran
Jess, Tine
Bhatt, Samir
Sazonovs, Aleksejs
author_facet Bechler-Speicher, Maya
Zerio, Andrea
Huri, Maor
Vestergaard, Marie Vibeke
Gilad-Bachrach, Ran
Jess, Tine
Bhatt, Samir
Sazonovs, Aleksejs
contents We introduce GMAN, a flexible, interpretable, and expressive framework that extends Graph Neural Additive Networks (GNANs) to learn from sets of sparse time-series data. GMAN represents each time-dependent trajectory as a directed graph and applies an enriched, more expressive GNAN to each graph. It allows users to control the interpretability-expressivity trade-off by grouping features and graphs to encode priors, and it provides feature, node, and graph-level interpretability. On real-world datasets, including mortality prediction from blood tests and fake-news detection, GMAN outperforms strong non-interpretable black-box baselines while delivering actionable, domain-aligned explanations.
format Preprint
id arxiv_https___arxiv_org_abs_2509_23923
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph Mixing Additive Networks
Bechler-Speicher, Maya
Zerio, Andrea
Huri, Maor
Vestergaard, Marie Vibeke
Gilad-Bachrach, Ran
Jess, Tine
Bhatt, Samir
Sazonovs, Aleksejs
Machine Learning
Artificial Intelligence
We introduce GMAN, a flexible, interpretable, and expressive framework that extends Graph Neural Additive Networks (GNANs) to learn from sets of sparse time-series data. GMAN represents each time-dependent trajectory as a directed graph and applies an enriched, more expressive GNAN to each graph. It allows users to control the interpretability-expressivity trade-off by grouping features and graphs to encode priors, and it provides feature, node, and graph-level interpretability. On real-world datasets, including mortality prediction from blood tests and fake-news detection, GMAN outperforms strong non-interpretable black-box baselines while delivering actionable, domain-aligned explanations.
title Graph Mixing Additive Networks
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2509.23923