Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
This paper suggests and experimentally validates a novel regularization method to enhaned adversarial robustness of a neural network image classifier. The proposed method is carefully motivated and introduced and extensively validated. The authors claim improved computational efficiency while (mostly) achieving state of the art performance in terms of adversarial robustness. No theoretical analysis is provided. The reviewers appreciated the work. Some reviewers have pointed out weaknesses in the experimental setup, which the authors promised to clarify in the final version. The authors are encouraged to carefully take the reviewers comments and the commitments in their own responses into account when preparing the final version.