Are Graph Convolutional Networks Fully Exploiting Graph Structure?

06/06/2020
by   Davide Buffelli, et al.
0

Graph Convolutional Networks (GCNs) generalize the idea of deep convolutional networks to graphs, and achieve state-of-the-art results on many graph related tasks. GCNs rely on the graph structure to define an aggregation strategy where each node updates its representation by combining information from its neighbours. In this paper we formalize four levels of structural information injection, and use them to show that GCNs ignore important long-range dependencies embedded in the overall topology of a graph. Our proposal includes a novel regularization technique based on random walks with restart, called RWRReg, which encourages the network to encode long-range information into the node embeddings. RWRReg is further supported by our theoretical analysis, which demonstrates that random walks with restart empower aggregation-based strategies (i.e., the Weisfeiler-Leman algorithm) with long-range information. We conduct an extensive experimental analysis studying the change in performance of several state-of-the-art models given by the four levels of structural information injection, on both transductive and inductive tasks. The results show that the lack of long-range structural information greatly affects performance on all considered models, and that the information extracted by random walks with restart, and exploited by RWRReg, gives an average accuracy improvement of more than 5% on all considered tasks.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/18/2023

RBA-GCN: Relational Bilevel Aggregation Graph Convolutional Network for Emotion Recognition

Emotion recognition in conversation (ERC) has received increasing attent...
research
02/13/2020

Geom-GCN: Geometric Graph Convolutional Networks

Message-passing neural networks (MPNNs) have been successfully applied t...
research
05/27/2022

Capturing Graphs with Hypo-Elliptic Diffusions

Convolutional layers within graph neural networks operate by aggregating...
research
06/09/2018

Representation Learning on Graphs with Jumping Knowledge Networks

Recent deep learning approaches for representation learning on graphs fo...
research
09/08/2019

Aspect-based Sentiment Classification with Aspect-specific Graph Convolutional Networks

Due to their inherent capability in semantic alignment of aspects and th...
research
11/22/2019

SWAG: Item Recommendations using Convolutions on Weighted Graphs

Recent advancements in deep neural networks for graph-structured data ha...
research
01/12/2023

A Network Science perspective of Graph Convolutional Networks: A survey

The mining and exploitation of graph structural information have been th...

Please sign up or login with your details

Forgot password? Click here to reset