NeurIPS 2020

Private Learning of Halfspaces: Simplifying the Construction and Reducing the Sample Complexity


Meta Review

There is a consensus among the reviewers that the paper provides new contribution to solving a fundamental problem of differentially private learning of half-spaces.