Creating generalizable downstream graph models with random projections
This addresses the challenge of graph model generalization for applications requiring cross-graph inference, though it appears incremental as it builds on existing random projection techniques.
The paper tackles the problem of enabling graph models to generalize across different graphs, rather than just unseen nodes within the same graph, by using random projections to compute isomorphism-invariant features. The result is a method that achieves competitive inference accuracy with minimal degradation while maintaining computational efficiency.
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.