Bayesian Robust Graph Contrastive Learning
This addresses robustness issues in graph neural networks for tasks like node classification, but it is incremental as it builds on existing contrastive learning and Bayesian methods.
The paper tackles the problem of noise in graph data degrading GNN performance by proposing Bayesian Robust Graph Contrastive Learning (BRGCL), an unsupervised method that iteratively estimates confident nodes and uses prototypical contrastive learning, achieving superior performance on public and large-scale benchmarks.
Graph Neural Networks (GNNs) have been widely used to learn node representations and with outstanding performance on various tasks such as node classification. However, noise, which inevitably exists in real-world graph data, would considerably degrade the performance of GNNs as the noise is easily propagated via the graph structure. In this work, we propose a novel and robust method, Bayesian Robust Graph Contrastive Learning (BRGCL), which trains a GNN encoder to learn robust node representations. The BRGCL encoder is a completely unsupervised encoder. Two steps are iteratively executed at each epoch of training the BRGCL encoder: (1) estimating confident nodes and computing robust cluster prototypes of node representations through a novel Bayesian nonparametric method; (2) prototypical contrastive learning between the node representations and the robust cluster prototypes. Experiments on public and large-scale benchmarks demonstrate the superior performance of BRGCL and the robustness of the learned node representations. The code of BRGCL is available at \url{https://github.com/BRGCL-code/BRGCL-code}.