23
1

Efficient Active Learning Halfspaces with Tsybakov Noise: A Non-convex Optimization Approach

Abstract

We study the problem of computationally and label efficient PAC active learning dd-dimensional halfspaces with Tsybakov Noise~\citep{tsybakov2004optimal} under structured unlabeled data distributions. Inspired by~\cite{diakonikolas2020learning}, we prove that any approximate first-order stationary point of a smooth nonconvex loss function yields a halfspace with a low excess error guarantee. In light of the above structural result, we design a nonconvex optimization-based algorithm with a label complexity of O~(d(1ϵ)86α3α1)\tilde{O}(d (\frac{1}{\epsilon})^{\frac{8-6\alpha}{3\alpha-1}})\footnote{In the main body of this work, we use O~(),Θ~()\tilde{O}(\cdot), \tilde{\Theta}(\cdot) to hide factors of the form \polylog(d,1ϵ,1δ)\polylog(d, \frac{1}{\epsilon}, \frac{1}{\delta})}, under the assumption that the Tsybakov noise parameter α(13,1]\alpha \in (\frac13, 1], which narrows down the gap between the label complexities of the previously known efficient passive or active algorithms~\citep{diakonikolas2020polynomial,zhang2021improved} and the information-theoretic lower bound in this setting.

View on arXiv
Comments on this paper