Learning Discrete Structures for Graph Neural Networks

28 Mar 2019Luca FranceschiMathias NiepertMassimiliano PontilXiao He

Graph neural networks (GNNs) are a popular class of machine learning models whose major advantage is their ability to incorporate a sparse and discrete dependency structure between data points. Unfortunately, GNNs can only be used when such a graph-structure is available... (read more)

PDF Abstract
Task Dataset Model Metric name Metric value Global rank Compare
Node Classification Citeseer LDS-GNN Accuracy 75.0% # 4
Node Classification CiteSeer with Public Split: fixed 20 nodes per class LDS-GNN Accuracy 75.0 # 1
Node Classification Cora LDS-GNN Accuracy 84.1% # 7
Node Classification Cora: fixed 20 node per class LDS-GNN Accuracy 84.1 # 1
Node Classification Cora with Public Split: fixed 20 nodes per class LDS-GNN Accuracy 84.1 # 2