Keep It Simple: Graph Autoencoders Without Graph Convolutional Networks

Abstract

Graph autoencoders (AE) and variational autoencoders (VAE) recently emerged as powerful node embedding methods, with promising performances on challenging tasks such as link prediction and node clustering. Graph AE, VAE and most of their extensions rely on graph convolutional networks (GCN) to learn vector space representations of nodes. In this paper, we propose to replace the GCN encoder by a simple linear model w.r.t. the adjacency matrix of the graph. For the two aforementioned tasks, we empirically show that this approach consistently reaches competitive performances w.r.t. GCN-based models for numerous real-world graphs, including the widely used Cora, Citeseer and Pubmed citation networks that became the de facto benchmark datasets for evaluating graph AE and VAE. This result questions the relevance of repeatedly using these three datasets to compare complex graph AE and VAE models. It also emphasizes the effectiveness of simple node encoding schemes for many real-world applications.

NeurIPS 2019 Graph Representation Learning Workshop

References in corpus (10)

• Auto-Encoding Variational Bayes
• Inductive Representation Learning on Large Graphs
• Spectral Networks and Locally Connected Networks on Graphs
• Simplifying Graph Convolutional Networks
• Graph Convolutional Matrix Completion
• Variational Graph Auto-Encoders
• Constrained Graph Variational Autoencoders for Molecule Design
• Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders
• RWR-GAE: Random Walk Regularization for Graph Auto Encoders
• Multi-Task Graph Autoencoders

Cited by in corpus (9)

• Local Augmentation for Graph Neural Networks
• Node Embeddings and Exact Low-Rank Representations of Complex Networks
• From Local Structures to Size Generalization in Graph Neural Networks
• A Review of Graph-Powered Data Quality Applications for IoT Monitoring Sensor Networks
• Graph-based State Representation for Deep Reinforcement Learning
• Bipartite Link Prediction based on Topological Features via 2-hop Path
• Thematic recommendations on knowledge graphs using multilayer networks
• Anomaly Detection for Aggregated Data Using Multi-Graph Autoencoder
• Effective Model Integration Algorithm for Improving Link and Sign Prediction in Complex Networks