Fishnets: Information-Optimal, Scalable Aggregation for Sets and Graphs
This addresses a key bottleneck in set-based and graph learning for applications in deep learning and network science, offering a scalable and robust alternative to existing methods.
The authors tackled the problem of learning informative embeddings for sets and graphs by proposing Fishnets, an information-optimal aggregation strategy, which achieved state-of-the-art performance on ogbn-protein data with fewer parameters and faster training.
Set-based learning is an essential component of modern deep learning and network science. Graph Neural Networks (GNNs) and their edge-free counterparts Deepsets have proven remarkably useful on ragged and topologically challenging datasets. The key to learning informative embeddings for set members is a specified aggregation function, usually a sum, max, or mean. We propose Fishnets, an aggregation strategy for learning information-optimal embeddings for sets of data for both Bayesian inference and graph aggregation. We demonstrate that i) Fishnets neural summaries can be scaled optimally to an arbitrary number of data objects, ii) Fishnets aggregations are robust to changes in data distribution, unlike standard deepsets, iii) Fishnets saturate Bayesian information content and extend to regimes where MCMC techniques fail and iv) Fishnets can be used as a drop-in aggregation scheme within GNNs. We show that by adopting a Fishnets aggregation scheme for message passing, GNNs can achieve state-of-the-art performance versus architecture size on ogbn-protein data over existing benchmarks with a fraction of learnable parameters and faster training time.