Accurate Node Feature Estimation with Structured Variational Graph Autoencoder

TL;DR

This paper introduces SVGA, a novel structured variational graph autoencoder that accurately estimates missing node features by leveraging probabilistic inference and graph structure, outperforming existing methods.

Contribution

The paper presents SVGA, a new approach that applies structured variational inference with Gaussian Markov random fields for improved high-dimensional feature estimation.

Findings

Achieves state-of-the-art performance on real datasets.

Effectively models prior distributions using graph structure.

Balances representation power with regularization to prevent overfitting.

Abstract

Given a graph with partial observations of node features, how can we estimate the missing features accurately? Feature estimation is a crucial problem for analyzing real-world graphs whose features are commonly missing during the data collection process. Accurate estimation not only provides diverse information of nodes but also supports the inference of graph neural networks that require the full observation of node features. However, designing an effective approach for estimating high-dimensional features is challenging, since it requires an estimator to have large representation power, increasing the risk of overfitting. In this work, we propose SVGA (Structured Variational Graph Autoencoder), an accurate method for feature estimation. SVGA applies strong regularization to the distribution of latent variables by structured variational inference, which models the prior of variables as Gaussian Markov random field based on the graph structure. As a result, SVGA combines the advantages of probabilistic inference and graph neural networks, achieving state-of-the-art performance in real datasets.