Training Graph Neural Networks by Graphon Estimation

TL;DR

This paper introduces a novel GNN training method using graphon estimation and resampling, which reduces over-smoothing and improves performance with minimal tuning, demonstrated through extensive numerical experiments.

Contribution

The paper presents a new GNN training framework based on graphon estimation and resampling, addressing over-smoothing effectively with a simple and efficient approach.

Findings

Outperforms existing over-smoothing mitigation methods in many cases

Requires minimal additional tuning during training

Demonstrates competitive or superior results in numerical experiments

Abstract

In this work, we propose to train a graph neural network via resampling from a graphon estimate obtained from the underlying network data. More specifically, the graphon or the link probability matrix of the underlying network is first obtained from which a new network will be resampled and used during the training process at each layer. Due to the uncertainty induced from the resampling, it helps mitigate the well-known issue of over-smoothing in a graph neural network (GNN) model. Our framework is general, computationally efficient, and conceptually simple. Another appealing feature of our method is that it requires minimal additional tuning during the training process. Extensive numerical results show that our approach is competitive with and in many cases outperform the other over-smoothing reducing GNN training methods.