Agnostic Learnability of Halfspaces via Logistic LossDownload PDFOpen Website

2022 (modified: 24 Apr 2023)ICML 2022Readers: Everyone
Abstract: We investigate approximation guarantees provided by logistic regression for the fundamental problem of agnostic learning of homogeneous halfspaces. Previously, for a certain broad class of “well-be...
0 Replies

Loading