EdgeSHAPer: Bond-Centric Shapley Value-Based Explanation Method for Graph Neural Networks
A Mastropietro, G Pasculli, C Feldmann… - iScience, 2022 - oak.novartis.com
Graph neural networks (GNNs) are becoming increasingly popular for many deep machine
learning (ML) applications in science. By recursively propagating neural signals along the …
learning (ML) applications in science. By recursively propagating neural signals along the …
[PDF][PDF] EdgeSHAPer: Bond-centric Shapley value-based explanation method for graph neural networks
G Pasculli, R Feldmann, J Bajorath - iris.uniroma1.it
Graph neural networks (GNNs) recursively propagate signals along the edges of an input
graph, integrate node feature information with graph structure, and learn object …
graph, integrate node feature information with graph structure, and learn object …