A Degeneracy Framework for Scalable Graph Autoencoders
This addresses the scalability issue for researchers and practitioners working with large-scale graph data, though it is incremental as it builds on existing graph AE/VAE methods.
The paper tackles the scalability problem of graph autoencoders and variational autoencoders by introducing a framework that uses graph degeneracy to train on dense subsets of nodes, achieving competitive results on large graphs with millions of nodes and edges while improving training speed.
In this paper, we present a general framework to scale graph autoencoders (AE) and graph variational autoencoders (VAE). This framework leverages graph degeneracy concepts to train models only from a dense subset of nodes instead of using the entire graph. Together with a simple yet effective propagation mechanism, our approach significantly improves scalability and training speed while preserving performance. We evaluate and discuss our method on several variants of existing graph AE and VAE, providing the first application of these models to large graphs with up to millions of nodes and edges. We achieve empirically competitive results w.r.t. several popular scalable node embedding methods, which emphasizes the relevance of pursuing further research towards more scalable graph AE and VAE.