Reproduce “Graph Information Bottleneck” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Tailin Wu, Hongyu Ren, Pan Li, Jure Leskovec

Graph Information Bottleneck (GIB) trains graph neural networks to preserve label-relevant information while limiting information retained from node features and graph structure. It uses a local-dependence assumption and samples neighbors and node representations. On targeted Pubmed Nettack tests with one perturbation, GIB-Cat and GIB-Bern exceed GAT in evasive and poisoning accuracy; results cover selected targets and specific protocols, not general robustness.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.