Nearly Tight Bounds for Robust Proper Learning of Halfspaces with a Margin
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 32 (NIPS 2019), pp. 10473-10484, 2019.
exponential time hypothesisunit ball
We study the problem of properly learning large margin halfspaces in the agnostic PAC model. In more detail, we study the complexity of properly learning d-dimensional halfspaces on the unit ball within misclassification error alpha. OPT gamma+epsilon where OPT gamma is the optimal gamma-margin error rate and alpha >= 1 is the approximati...More
PPT (Upload PPT)