First- and second-order high probability complexity bounds for trust-region methods with noisy oracles

DOI10.1007/S10107-023-01999-5zbMATH Open1547.65056MaRDI QIDQ6608030FDOQ6608030

Authors: Liyuan Cao, Albert S. Berahas, Katya Scheinberg 

Publication date: 19 September 2024

Published in: Mathematical Programming. Series A. Series B (Search for Journal in Brave)

zbMATH Keywords

complexityboundsprobability

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonlinear programming (90C30) Derivative-free methods and methods using generalized derivatives (90C56)

Cites Work

• UOBYQA: unconstrained optimization by quadratic approximation
• High-dimensional probability. An introduction with applications in data science
• Sample size selection in optimization methods for machine learning
• Trust Region Methods
• Random gradient-free minimization of convex functions
• Introduction to Derivative-Free Optimization
• Benchmarking Derivative-Free Optimization Algorithms
• Recent advances in trust region algorithms
• Stochastic optimization using a trust-region method and random models
• Convergence of trust-region methods based on probabilistic models
• Computation of sparse low degree interpolating polynomials and their application to derivative-free optimization
• Cubic regularization of Newton method and its global performance
• On the Global Convergence of Trust Region Algorithms Using Inexact Gradient Information
• A stochastic line search method with expected complexity analysis
• Derivative-Free Optimization of Noisy Functions via Quasi-Newton Methods
• Global convergence rate analysis of unconstrained optimization methods based on probabilistic models
• On the Lagrange functions of quadratic models that are defined by interpolation^*
• Global Convergence Rate Analysis of a Generic Line Search Algorithm with Noise
• Complexity and global rates of trust-region methods based on probabilistic models

Cited In (2)

• A non-monotone trust-region method with noisy oracles and additional sampling
• A sequential quadratic programming method with high-probability complexity bounds for nonlinear equality-constrained stochastic optimization

This page was built for publication: First- and second-order high probability complexity bounds for trust-region methods with noisy oracles

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6608030)