Neighbor Contrastive Learning with Learnable Graph Augmentation for Robust Semi-Supervised Node Classification Under Structural

Recent years, graph contrastive learning (GCL), which aims to learn representations from unlabeled graphs, has made great progress. However, the existing GCL methods mostly adopt human-designed graph augmentations, which are sensitive to various graph datasets. In addition, the contrastive losses originally developed in computer vision have been directly applied to graph data, where the neighboring nodes are regarded as negatives and consequently pushed far apart from the anchor. However, this is contradictory with the homophily assumption of net-works that connected nodes often belong to the

Research goal: To what extent does neighbor contrastive learning with learnable graph augmentation improve robustness against structural noise in semi-supervised node classification tasks?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.7/10.