S-CGIB
PublicSubgraph-conditioned Graph Information Bottleneck (S-CGIB) is a novel architecture for pre-training Graph Neural Networks in molecular property prediction and developed by NS Lab, CUK based on pure PyTorch backend.
graphgraph-foundation-modelgraph-information-bottleneckgraph-neural-networksgraph-representation-learningmolecular-graph-learningmolecular-property-predictionpretrained-graph-modelself-supervised-graph-learning
Creat:2024-12-11T15:43:14
Update:2025-06-15T04:05:13
https://nslab-cuk.github.io/2024/12/19/SCGIB/
13
Stars
0
Stars Increase