Creating generalizable downstream graph models with random projections

TL;DR

This paper introduces a method using random projections to create graph features that generalize across different graphs, enabling accurate inference on unseen graphs with high efficiency.

Contribution

The paper proposes a novel approach employing random projections to generate isomorphism-invariant features for cross-graph generalization in graph representation learning.

Findings

Features enable inference on unseen graphs with minimal accuracy loss

Method maintains computational efficiency

Achieves relevance competitive results compared to existing approaches

Abstract

We investigate graph representation learning approaches that enable models to generalize across graphs: given a model trained using the representations from one graph, our goal is to apply inference using those same model parameters when given representations computed over a new graph, unseen during model training, with minimal degradation in inference accuracy. This is in contrast to the more common task of doing inference on the unseen nodes of the same graph. We show that using random projections to estimate multiple powers of the transition matrix allows us to build a set of isomorphism-invariant features that can be used by a variety of tasks. The resulting features can be used to recover enough information about the local neighborhood of a node to enable inference with relevance competitive to other approaches while maintaining computational efficiency.