Ego-GNNs: Exploiting Ego Structures in Graph Neural Networks

07/22/2021
by   Dylan Sandfelder, et al.
19

Graph neural networks (GNNs) have achieved remarkable success as a framework for deep learning on graph-structured data. However, GNNs are fundamentally limited by their tree-structured inductive bias: the WL-subtree kernel formulation bounds the representational capacity of GNNs, and polynomial-time GNNs are provably incapable of recognizing triangles in a graph. In this work, we propose to augment the GNN message-passing operations with information defined on ego graphs (i.e., the induced subgraph surrounding each node). We term these approaches Ego-GNNs and show that Ego-GNNs are provably more powerful than standard message-passing GNNs. In particular, we show that Ego-GNNs are capable of recognizing closed triangles, which is essential given the prominence of transitivity in real-world graphs. We also motivate our approach from the perspective of graph signal processing as a form of multiplex graph convolution. Experimental results on node classification using synthetic and real data highlight the achievable performance gains using this approach.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/15/2021

Neural Trees for Learning on Graphs

Graph Neural Networks (GNNs) have emerged as a flexible and powerful app...
research
04/07/2021

Theoretically Improving Graph Neural Networks via Anonymous Walk Graph Kernels

Graph neural networks (GNNs) have achieved tremendous success in graph m...
research
06/30/2021

Informed Machine Learning for Improved Similarity Assessment in Process-Oriented Case-Based Reasoning

Currently, Deep Learning (DL) components within a Case-Based Reasoning (...
research
05/30/2021

How effective are Graph Neural Networks in Fraud Detection for Network Data?

Graph-based Neural Networks (GNNs) are recent models created for learnin...
research
08/16/2023

Expressivity of Graph Neural Networks Through the Lens of Adversarial Robustness

We perform the first adversarial robustness study into Graph Neural Netw...
research
02/15/2022

Geometrically Equivariant Graph Neural Networks: A Survey

Many scientific problems require to process data in the form of geometri...
research
12/18/2022

Graph Neural Networks are Inherently Good Generalizers: Insights by Bridging GNNs and MLPs

Graph neural networks (GNNs), as the de-facto model class for representa...

Please sign up or login with your details

Forgot password? Click here to reset