NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:932
Title:ResNets Ensemble via the Feynman-Kac Formalism to Improve Natural and Robust Accuracies

Good paper, accept. Please add the following points included in the rebuttal to the final version: - inference time and memory consumption - additional results on the robustness of the proposal with natural training. - additional results about the robustness of the methods.