English

Acceleration with a Ball Optimization Oracle

Optimization and Control 2020-03-19 v1 Data Structures and Algorithms

Abstract

Consider an oracle which takes a point xx and returns the minimizer of a convex function ff in an 2\ell_2 ball of radius rr around xx. It is straightforward to show that roughly r1log1ϵr^{-1}\log\frac{1}{\epsilon} calls to the oracle suffice to find an ϵ\epsilon-approximate minimizer of ff in an 2\ell_2 unit ball. Perhaps surprisingly, this is not optimal: we design an accelerated algorithm which attains an ϵ\epsilon-approximate minimizer with roughly r2/3log1ϵr^{-2/3} \log \frac{1}{\epsilon} 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 \ell_\infty regression and achieving guarantees comparable to the state-of-the-art for p\ell_p regression.

Keywords

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