Acceleration with a Ball Optimization Oracle
Abstract
Consider an oracle which takes a point and returns the minimizer of a convex function in an ball of radius around . It is straightforward to show that roughly calls to the oracle suffice to find an -approximate minimizer of in an unit ball. Perhaps surprisingly, this is not optimal: we design an accelerated algorithm which attains an -approximate minimizer with roughly oracle queries, and give a matching lower bound. Further, we implement ball optimization oracles for functions with locally stable Hessians using a variant of Newton's method. The resulting algorithm applies to a number of problems of practical and theoretical import, improving upon previous results for logistic and regression and achieving guarantees comparable to the state-of-the-art for regression.
Cite
@article{arxiv.2003.08078,
title = {Acceleration with a Ball Optimization Oracle},
author = {Yair Carmon and Arun Jambulapati and Qijia Jiang and Yujia Jin and Yin Tat Lee and Aaron Sidford and Kevin Tian},
journal= {arXiv preprint arXiv:2003.08078},
year = {2020}
}
Comments
37 pages