search

Shen-Lab / GraphCL

[NeurIPS 2020] "Graph Contrastive Learning with Augmentations" by Yuning You, Tianlong Chen, Yongduo Sui, Ting Chen, Zhangyang Wang, Yang Shen
MIT License
548 stars 103 forks source link

Question about data augmentation #56

Open Struggle-Forever opened 2 years ago

Struggle-Forever commented 2 years ago

Question about data augmentation: When subgraph augmentation is performed, essentially only the enge_index is modified. This means that only the subgraph is aggregated for information propagation. What puzzles me is that when computing the graph level features, the pooling function involves the features of all nodes of the graph (selected subgraph nodes and unselected nodes), does this make sense? Isn't the pooling at this point only considering subgraphs?

yyou1996 commented 2 years ago

Hi @Struggle-Forever,

Please refer to an instantiation of the subgraph augmentation https://github.com/Shen-Lab/GraphCL/issues/24#issuecomment-839406281. We would drop the non-selected nodes via, e.g. data.x = data.x[idx_nondrop] where the pooling function is operated on.