Stopping rules for gradient methods for non-convex problems with additive noise in gradient

DOI10.1007/s10957-023-02245-wzbMath1522.90141arXiv2205.07544OpenAlexW4379010939MaRDI QIDQ6051170

Boris T. Polyak, Ilya A. Kuruzov, Fedor S. Stonyakin

Publication date: 19 September 2023

Published in: Journal of Optimization Theory and Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2205.07544

zbMATH Keywords

stopping ruleadaptive methodnon-convex optimizationPolyak-Łojasiewicz conditioninexact gradient

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonconvex programming, global optimization (90C26) Methods of reduced gradient type (90C52)

Related Items

Gradient-Type Methods for Optimization Problems with Polyak-Łojasiewicz Condition: Early Stopping and Adaptivity to Inexactness Parameter

Cites Work

• Unnamed Item
• First-order methods of smooth convex optimization with inexact oracle
• Universal gradient methods for convex optimization problems
• A stopping rule in iteration procedures for solving ill-posed problems
• Cubic regularization of Newton method and its global performance
• Smooth Optimization with Approximate Gradient
• New versions of Newton method: step-size choice, convergence domain and under-determined equations
• Fit without fear: remarkable mathematical phenomena of deep learning through the prism of interpolation

This page was built for publication: Stopping rules for gradient methods for non-convex problems with additive noise in gradient