A Universally Optimal Multistage Accelerated Stochastic Gradient Method
Optimization and Control
2019-10-29 v3 Machine Learning
Machine Learning
Abstract
We study the problem of minimizing a strongly convex, smooth function when we have noisy estimates of its gradient. We propose a novel multistage accelerated algorithm that is universally optimal in the sense that it achieves the optimal rate both in the deterministic and stochastic case and operates without knowledge of noise characteristics. The algorithm consists of stages that use a stochastic version of Nesterov's method with a specific restart and parameters selected to achieve the fastest reduction in the bias-variance terms in the convergence rate bounds.
Cite
@article{arxiv.1901.08022,
title = {A Universally Optimal Multistage Accelerated Stochastic Gradient Method},
author = {Necdet Serhat Aybat and Alireza Fallah and Mert Gurbuzbalaban and Asuman Ozdaglar},
journal= {arXiv preprint arXiv:1901.08022},
year = {2019}
}
Comments
33rd Conference on Neural Information Processing Systems (NeurIPS 2019)