NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:4956
Title:GNNExplainer: Generating Explanations for Graph Neural Networks

The reviewers agreed that this paper presents a valuable contribution for explaining GNNs; they appreciated the quality of the writing, the overall motivation of interpretability of the models, and the strength of the empirical results. The primary remaining shortcomings that the reviewers mentioned in the reviews should be addressed as described in the response, such as expanding explanations that describe multiple edges, the significance of the hyper-parameters, description of synthetic datasets, and the additional experiments.