Trust region-type method under inexact gradient and inexact Hessian with convergence analysis

DOI10.12286/jssx.j2022-0960MaRDI QIDQ6665208

Dele Ma, Xiangmei Wang

Publication date: 17 January 2025

Published in: Mathematica Numerica Sinica (Search for Journal in Brave)

zbMATH Keywords

trust-region algorithmmachine learningsampling approximationadaptive cubic regularization algorithm

Mathematics Subject Classification ID

Computer-aided design (modeling of curves and surfaces) (65D17)

Cites Work

• Unnamed Item
• A trust region algorithm with a worst-case iteration complexity of \(\mathcal{O}(\epsilon ^{-3/2})\) for nonconvex optimization
• Adaptive cubic regularisation methods for unconstrained optimization. I: Motivation, convergence and numerical results
• Complexity bounds for second-order optimality in unconstrained optimization
• Sub-sampled Newton methods
• Newton-type methods for non-convex optimization under inexact Hessian information
• Cubic regularization of Newton method and its global performance
• On the use of iterative methods in cubic regularization for unconstrained optimization
• On the worst-case complexity of nonlinear stepsize control algorithms for convex unconstrained optimization
• Newton Sketch: A Near Linear-Time Optimization Algorithm with Linear-Quadratic Convergence
• Iterative Methods for Finding a Trust-region Step
• Newton’s Method with a Model Trust Region Modification
• Solving the Trust-Region Subproblem using the Lanczos Method
• A Stochastic Approximation Method

This page was built for publication: Trust region-type method under inexact gradient and inexact Hessian with convergence analysis