Recovering Missing Node Features with Local Structure-based Embeddings

TL;DR

This paper introduces a framework for recovering missing node features in graphs by leveraging local structure-based embeddings, improving feature estimation and downstream classification performance.

Contribution

It proposes a novel method that uses local graph structure and a graph autoencoder to estimate missing node features across multiple graphs.

Findings

Accurately estimates missing node features using local structure embeddings.

Enhances downstream graph classification performance.

Demonstrates the importance of structure-feature relationship in graph learning.

Abstract

Node features bolster graph-based learning when exploited jointly with network structure. However, a lack of nodal attributes is prevalent in graph data. We present a framework to recover completely missing node features for a set of graphs, where we only know the signals of a subset of graphs. Our approach incorporates prior information from both graph topology and existing nodal values. We demonstrate an example implementation of our framework where we assume that node features depend on local graph structure. Missing nodal values are estimated by aggregating known features from the most similar nodes. Similarity is measured through a node embedding space that preserves local topological features, which we train using a Graph AutoEncoder. We empirically show not only the accuracy of our feature estimation approach but also its value for downstream graph classification. Our success embarks on and implies the need to emphasize the relationship between node features and graph structure in graph-based learning.