Discriminability of Single-Layer Graph Neural Networks

10/17/2020
by   Samuel Pfrommer, et al.
0

Network data can be conveniently modeled as a graph signal, where data values are assigned to the nodes of a graph describing the underlying network topology. Successful learning from network data requires methods that effectively exploit this graph structure. Graph neural networks (GNNs) provide one such method and have exhibited promising performance on a wide range of problems. Understanding why GNNs work is of paramount importance, particularly in applications involving physical networks. We focus on the property of discriminability and establish conditions under which the inclusion of pointwise nonlinearities to a stable graph filter bank leads to an increased discriminative capacity for high-eigenvalue content. We define a notion of discriminability tied to the stability of the architecture, show that GNNs are at least as discriminative as linear graph filter banks, and characterize the signals that cannot be discriminated by either.

READ FULL TEXT
research
03/08/2020

Graphs, Convolutions, and Neural Networks

Network data can be conveniently modeled as a graph signal, where data v...
research
10/21/2019

Stability of Graph Neural Networks to Relative Perturbations

Graph neural networks (GNNs), consisting of a cascade of layers applying...
research
07/08/2023

Manifold Filter-Combine Networks

We introduce a class of manifold neural networks (MNNs) that we call Man...
research
08/26/2021

Web Image Context Extraction with Graph Neural Networks and Sentence Embeddings on the DOM tree

Web Image Context Extraction (WICE) consists in obtaining the textual in...
research
05/11/2019

Stability Properties of Graph Neural Networks

Data stemming from networks exhibit an irregular support, whereby each d...
research
06/11/2020

Graph Neural Networks for Motion Planning

This paper investigates the feasibility of using Graph Neural Networks (...
research
10/22/2020

Stability of Algebraic Neural Networks to Small Perturbations

Algebraic neural networks (AlgNNs) are composed of a cascade of layers e...

Please sign up or login with your details

Forgot password? Click here to reset