GNNExplainer: Generating Explanations for Graph Neural Networks
PulseAugur coverage of GNNExplainer: Generating Explanations for Graph Neural Networks — every cluster mentioning GNNExplainer: Generating Explanations for Graph Neural Networks across labs, papers, and developer communities, ranked by signal.
-
GNN explanation methods reveal disease signatures in biological networks
Researchers have evaluated four popular explanation methods for graph neural networks (GNNs) to understand their effectiveness in identifying disease-associated structures within biological networks. Using synthetic dat…
-
New attack reconstructs private graph data from GNN explanations
Researchers have developed a new attack called PRIVX that can reconstruct hidden graph structures from differentially private Graph Neural Network (GNN) explanations. The attack exploits the Gaussian differential privac…
-
AI researchers compare explainability methods for jet tagging in particle physics
Researchers have developed and compared three explainable AI (XAI) methods—GNNExplainer, GNNShap, and GradCAM—to understand the predictions of graph neural networks used in jet tagging at the Large Hadron Collider. The …