NeurIPS 2020

DiffGCN: Graph Convolutional Networks via Differential Operators and Algebraic Multigrid Pooling

Meta Review

The paper proposes a novel approach for a relevant problem and the differential operators are interesting/useful for the community. The paper lacks clarity and technical details should be better described in the camera ready version. After the rebuttal, the reviewers agree that the approach is sufficiently interesting, novel and performs comparable/superior to the state of the art and should be accepted.