NeurIPS 2020

The Complexity of Adversarially Robust Proper Learning of Halfspaces with Agnostic Noise

Meta Review

The four reviewers in their initial reviews all recommended accepting. On the positive side, the topic is timely and the paper introduces techniques that might be of interest to other researchers in the field. On the negative side, the achieved results feel somewhat incremental considering recent related work. After the author reply and discussion, the reviewers are even more strongly unanimous about accepting the paper.