FastGAE: Scalable Graph Autoencoders with Stochastic Subgraph Decoding
This addresses the problem of scaling graph AE and VAE for large real-world graphs, which is incremental as it builds on existing methods with a novel decoding scheme.
The paper tackled the scalability issues of graph autoencoders (AE) and variational autoencoders (VAE) by introducing FastGAE, a framework that uses stochastic subgraph decoding to speed up training on large graphs with millions of nodes and edges, while preserving or improving performance and outperforming existing approaches by a wide margin.
Graph autoencoders (AE) and variational autoencoders (VAE) are powerful node embedding methods, but suffer from scalability issues. In this paper, we introduce FastGAE, a general framework to scale graph AE and VAE to large graphs with millions of nodes and edges. Our strategy, based on an effective stochastic subgraph decoding scheme, significantly speeds up the training of graph AE and VAE while preserving or even improving performances. We demonstrate the effectiveness of FastGAE on various real-world graphs, outperforming the few existing approaches to scale graph AE and VAE by a wide margin.