Salvato in:
Dettagli Bibliografici
Autori principali: Abate, Carlo, Bianchi, Filippo Maria
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2409.05100
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929692923133952
author Abate, Carlo
Bianchi, Filippo Maria
author_facet Abate, Carlo
Bianchi, Filippo Maria
contents We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach works well on any kind of graph topology and can find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, trainable end-to-end, and particularly suitable for downstream tasks on heterophilic graphs.
format Preprint
id arxiv_https___arxiv_org_abs_2409_05100
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks
Abate, Carlo
Bianchi, Filippo Maria
Machine Learning
We propose a novel approach to compute the MAXCUT in attributed graphs, i.e., graphs with features associated with nodes and edges. Our approach works well on any kind of graph topology and can find solutions that jointly optimize the MAXCUT along with other objectives. Based on the obtained MAXCUT partition, we implement a hierarchical graph pooling layer for Graph Neural Networks, which is sparse, trainable end-to-end, and particularly suitable for downstream tasks on heterophilic graphs.
title MaxCutPool: differentiable feature-aware Maxcut for pooling in graph neural networks
topic Machine Learning
url https://arxiv.org/abs/2409.05100