flyingdoog / PGExplainer

Parameterized Explainer for Graph Neural Network
128 stars 15 forks source link

About the number of samples K in the paper. #7

Closed guyguygang closed 3 years ago

guyguygang commented 3 years ago

Hi,

Really cool project. When I compared this implementation with the one decribed in the paper, I found out that this code does not sample K times for each node's subgraph. Is there any reason for it and will it leads to a high variance for a certain node?

Looking forward to your response!

Thanks!

flyingdoog commented 3 years ago

We just set k=1 due to its empirically good performance. Theoretically large k leads to smaller variance although we didn’t empirically verify it.