User: Guest  Login
Document type:
Konferenzbeitrag
Contribution type:
Poster
Author(s):
Frederik Diehl, Thomas Brunner, Michael Truong Le and Alois Knoll
Title:
Towards Graph Pooling by Edge Contraction
Abstract:
Graph Neural Networks (GNNs) research has concentrated on improving convolutional layers, with little attention paid to developing graph poolinglayers. Yet pooling layers can enable GNNs toreason over abstracted groups of nodes instead ofsingle nodes, thus increasing their generalization potential. To close this gap, we propose a graph pooling layer relying on the notion of edge con-traction: EdgePool learns a localized and sparse pooling transform. We evaluate it on four datasets, finding that...     »
Book / Congress title:
ICML 2019 Workshop on Learning and Reasoning with Graph-Structured Data
Date of publication:
15.06.2019
Year:
2019
 BibTeX