Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kelesis, Dimitrios, Fotakis, Dimitris, Paliouras, Georgios
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916463117336576
author Kelesis, Dimitrios
Fotakis, Dimitris
Paliouras, Georgios
author_facet Kelesis, Dimitrios
Fotakis, Dimitris
Paliouras, Georgios
contents In this work, we generalize the ideas of Kaiming initialization to Graph Neural Networks (GNNs) and propose a new scheme (G-Init) that reduces oversmoothing, leading to very good results in node and graph classification tasks. GNNs are commonly initialized using methods designed for other types of Neural Networks, overlooking the underlying graph topology. We analyze theoretically the variance of signals flowing forward and gradients flowing backward in the class of convolutional GNNs. We then simplify our analysis to the case of the GCN and propose a new initialization method. Our results indicate that the new method (G-Init) reduces oversmoothing in deep GNNs, facilitating their effective use. Experimental validation supports our theoretical findings, demonstrating the advantages of deep networks in scenarios with no feature information for unlabeled nodes (i.e., ``cold start'' scenario).
format Preprint
id arxiv_https___arxiv_org_abs_2410_23830
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks
Kelesis, Dimitrios
Fotakis, Dimitris
Paliouras, Georgios
Machine Learning
In this work, we generalize the ideas of Kaiming initialization to Graph Neural Networks (GNNs) and propose a new scheme (G-Init) that reduces oversmoothing, leading to very good results in node and graph classification tasks. GNNs are commonly initialized using methods designed for other types of Neural Networks, overlooking the underlying graph topology. We analyze theoretically the variance of signals flowing forward and gradients flowing backward in the class of convolutional GNNs. We then simplify our analysis to the case of the GCN and propose a new initialization method. Our results indicate that the new method (G-Init) reduces oversmoothing in deep GNNs, facilitating their effective use. Experimental validation supports our theoretical findings, demonstrating the advantages of deep networks in scenarios with no feature information for unlabeled nodes (i.e., ``cold start'' scenario).
title Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks
topic Machine Learning
url https://arxiv.org/abs/2410.23830