Skip to content

Revisiting the Polyak Step Size

Authors
Elad Hazan, Sham Kakade
Year
2019
Preprint
Source
arXiv
Identifiers
arXiv 1905.00313

Abstract

This paper revisits the Polyak step size schedule for convex optimization problems, proving that a simple variant of it simultaneously attains near optimal convergence rates for the gradient descent algorithm, for all ranges of strong convexity, smoothness, and Lipschitz parameters, without a-priory knowledge of these parameters.

Knowledge Base

Map Tree Search

External

Document arXiv Abstract arXiv PDF arXiv HTML Google Scholar OpenAlex

Tags

Convex optimization Gradient descent Optimization

Similar Papers