Chenhui Deng

Hi, we cannot directly set the coarsen level, but you can try a large reduction ratio first and then to see the relationship between reduction ratio and coarsen level via...

Hi, Simple coarsening adopts a similar idea of LAMG. Gauss-Seidel (GS) iteration is just one way to smooth node embedding. For simple coarsening, we use low-pass graph filter instead of...

As shown in our paper (specifically, Equation (1)), we first generate random vector for each node and then apply smoothing function to smooth it. The smoothed vector mainly consists of...

Hi, Thanks for your interest! Can you please elaborate more about your question? Are you using GraphZoom to obtain the node embeddings? Which dataset are you using? In addition, can...

We have never used l2 distance and the kernel matrix that you mentioned for the label prediction. Have you tried "Logistic Regression" for label prediction after obtaining node embeddings? If...

Hi, The table you showed is the fusion step. Can you please also show me the LAMG table during the coarsening phase? What is the reduction ratio you are using...

Thanks for the information. For the GraphSAGE baseline, are you using the same code in GraphZoom by disabling fusion, coarsening, and refinement? If so, would you mind sharing this dataset...

I see. In this case I think the main reason may be that the default hyperparameters of GraphSAGE in our code is not as good as the GraphSAGE code you...

I would suggest you to tune "max_total_steps", "learning rate", "neg_sample_size", and "hidden dimension". Btw, you are using the **unsupervised** version (instead of the supervised version) of GraphSAGE from stellargraph, right?

Good to know you find the reason. Note that GraphZoom is just a framework and you can plug in whatever graph embedding model you want. If there is another learning-based...