Rethinking Kernel Methods for Node Representation Learning on Graphs

Open in new window