Hierarchical and Unsupervised Graph Representation Learning with Loukas's Coarsening
This addresses the problem of learning graph embeddings without labels for researchers in graph machine learning, though it appears incremental as it builds on existing coarsening and mutual information techniques.
The authors tackled unsupervised graph representation learning for attributed graphs by developing an inductive, hierarchical algorithm that combines graph coarsening with mutual information maximization. They demonstrated competitive performance with state-of-the-art methods on standard classification benchmarks.
We propose a novel algorithm for unsupervised graph representation learning with attributed graphs. It combines three advantages addressing some current limitations of the literature: i) The model is inductive: it can embed new graphs without re-training in the presence of new data; ii) The method takes into account both micro-structures and macro-structures by looking at the attributed graphs at different scales; iii) The model is end-to-end differentiable: it is a building block that can be plugged into deep learning pipelines and allows for back-propagation. We show that combining a coarsening method having strong theoretical guarantees with mutual information maximization suffices to produce high quality embeddings. We evaluate them on classification tasks with common benchmarks of the literature. We show that our algorithm is competitive with state of the art among unsupervised graph representation learning methods.