DeepAI AI Chat
Log In Sign Up

OOD Link Prediction Generalization Capabilities of Message-Passing GNNs in Larger Test Graphs

05/30/2022
by   Yangze Zhou, et al.
0

This work provides the first theoretical study on the ability of graph Message Passing Neural Networks (gMPNNs) – such as Graph Neural Networks (GNNs) – to perform inductive out-of-distribution (OOD) link prediction tasks, where deployment (test) graph sizes are larger than training graphs. We first prove non-asymptotic bounds showing that link predictors based on permutation-equivariant (structural) node embeddings obtained by gMPNNs can converge to a random guess as test graphs get larger. We then propose a theoretically-sound gMPNN that outputs structural pairwise (2-node) embeddings and prove non-asymptotic bounds showing that, as test graphs grow, these embeddings converge to embeddings of a continuous function that retains its ability to predict links OOD. Empirical results on random graphs show agreement with our theoretical results.

READ FULL TEXT

page 1

page 2

page 3

page 4

06/20/2022

Two-Dimensional Weisfeiler-Lehman Graph Neural Networks for Link Prediction

Link prediction is one important application of graph neural networks (G...
02/01/2022

Stability and Generalization Capabilities of Message Passing Graph Neural Networks

Message passing neural networks (MPNN) have seen a steep rise in popular...
09/05/2020

A Simple and General Graph Neural Network with Stochastic Message Passing

Graph neural networks (GNNs) are emerging machine learning models on gra...
10/01/2019

On the Equivalence between Node Embeddings and Structural Graph Representations

This work provides the first unifying theoretical framework for node emb...
06/11/2020

Pointer Graph Networks

Graph neural networks (GNNs) are typically applied to static graphs that...
10/09/2021

Towards Open-World Feature Extrapolation: An Inductive Graph Learning Approach

We target open-world feature extrapolation problem where the feature spa...
06/15/2022

DiffWire: Inductive Graph Rewiring via the Lovász Bound

Graph Neural Networks (GNNs) have been shown to achieve competitive resu...