Stochastic analysis of an adaptive cubic regularization method under inexact gradient evaluations and dynamic Hessian accuracy

DOI10.1080/02331934.2021.1892104zbMath1483.65091arXiv2001.10827OpenAlexW3133851861MaRDI QIDQ5034938

Gianmarco Gurioli, Stefania Bellavia

Publication date: 21 February 2022

Published in: Optimization (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2001.10827

zbMATH Keywords

worst-case complexity analysisfinite-sum minimizationadaptive cubic regularization methodsinexact derivatives evaluationsstochastic nonconvex optimization

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonlinear programming (90C30) Methods of quasi-Newton type (90C53)

Related Items (3)

Trust-region algorithms: probabilistic complexity and intrinsic noise with applications to subsampling techniques ⋮ The impact of noise on evaluation complexity: the deterministic trust-region case ⋮ Linesearch Newton-CG methods for convex optimization with noise

Cites Work

• Unnamed Item
• Worst-case evaluation complexity for unconstrained nonlinear optimization using high-order regularized models
• Adaptive cubic regularisation methods for unconstrained optimization. I: Motivation, convergence and numerical results
• Adaptive cubic regularisation methods for unconstrained optimization. II: Worst-case function- and derivative-evaluation complexity
• Complexity bounds for second-order optimality in unconstrained optimization
• Global convergence rate analysis of unconstrained optimization methods based on probabilistic models
• Stochastic optimization using a trust-region method and random models
• Newton-type methods for non-convex optimization under inexact Hessian information
• A note on inexact gradient and Hessian conditions for cubic regularized Newton's method
• Cubic regularization of Newton method and its global performance
• On the use of iterative methods in cubic regularization for unconstrained optimization
• On the Oracle Complexity of First-Order and Derivative-Free Algorithms for Smooth Nonconvex Minimization
• On the Complexity of Steepest Descent, Newton's and Regularized Newton's Methods for Nonconvex Unconstrained Optimization Problems
• Optimization Methods for Large-Scale Machine Learning
• An adaptive cubic regularization algorithm for nonconvex optimization with convex constraints and its function-evaluation complexity
• Adaptive cubic regularization methods with dynamic inexact Hessian information and applications to finite-sum minimization
• An investigation of Newton-Sketch and subsampled Newton methods
• A Stochastic Line Search Method with Expected Complexity Analysis
• Iteration and evaluation complexity for the minimization of functions whose computation is intrinsically inexact
• Adaptive Regularization Algorithms with Inexact Evaluations for Nonconvex Optimization
• An Introduction to Matrix Concentration Inequalities

This page was built for publication: Stochastic analysis of an adaptive cubic regularization method under inexact gradient evaluations and dynamic Hessian accuracy