A generalized worst-case complexity analysis for non-monotone line searches
From MaRDI portal
Publication:2028039
Abstract: We study the worst-case complexity of a non-monotone line search framework that covers a wide variety of known techniques published in the literature. In this framework, the non-monotonicity is controlled by a sequence of nonnegative parameters. We obtain complexity bounds to achieve approximate first-order optimality even when this sequence is not summable.
In this paper , the authors consider the worst-case complexity of a generalized version of the non-monotone line search framework proposed in [\textit{E. W. Sachs} and \textit{S. M. Sachs}, Control Cybern. 40, No. 4, 1059--1075 (2011; Zbl 1318.49059)] for smooth unconstrained optimization problems. This framework is built upon a generalized Armijo rule in which the level of non-monotonicity is controlled by a sequence of non-negative real parameters. The authors obtain complexity bounds to achieve approximate first-order optimality even when this sequence is not summable.
Recommendations
- On the worst-case evaluation complexity of non-monotone line search algorithms
- Worst-case evaluation complexity of non-monotone gradient-related algorithms for unconstrained optimization
- On the nonmonotone line search
- Nonmonotone line searches for optimization algorithms
- On the nonmonotonicity degree of nonmonotone line searches
Cites work
- A class of nonmonotone armijo-type line search method for unconstrained optimization
- A derivative-free line search and global convergence of Broyden-like method for nonlinear equations
- A line-search algorithm inspired by the adaptive cubic regularization framework and complexity analysis
- A new nonmonotone trust region method for unconstrained optimization equipped by an efficient adaptive radius
- A Nonmonotone Line Search Technique and Its Application to Unconstrained Optimization
- A Nonmonotone Line Search Technique for Newton’s Method
- A nonmonotone trust region method based on nonincreasing technique of weighted average of the successive function values
- 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. II: Worst-case function- and derivative-evaluation complexity
- Adaptive cubic regularization methods with dynamic inexact Hessian information and applications to finite-sum minimization
- An inexact line search approach using modified nonmonotone strategy for unconstrained optimization
- ARC\(_q\): a new adaptive regularization by cubics
- Benchmarking optimization software with performance profiles.
- Complexity analysis of second-order line-search algorithms for smooth nonconvex optimization
- Cubic regularization of Newton method and its global performance
- Cubic-regularization counterpart of a variable-norm trust-region method for unconstrained minimization
- Equation of state calculations by fast computing machines
- Generalized descent for global optimization
- scientific article; zbMATH DE number 3790208 (Why is no real title available?)
- Hybrid spectral gradient method for the unconstrained minimization problem
- Introductory lectures on convex optimization. A basic course.
- Lectures on convex optimization
- Newton-type methods for non-convex optimization under inexact Hessian information
- Nonlinear stepsize control algorithms: complexity bounds for first- and second-order optimality
- Nonmonotone line searches for optimization algorithms
- On efficiency of nonmonotone Armijo-type line searches
- On the complexity of steepest descent, Newton's and regularized Newton's methods for nonconvex unconstrained optimization problems
- On the convergence and worst-case complexity of trust-region and regularization methods for unconstrained optimization
- On the nonmonotonicity degree of nonmonotone line searches
- On the worst-case evaluation complexity of non-monotone line search algorithms
- Optimization by simulated annealing
- Optimization theory and methods. Nonlinear programming
- Recursive Trust-Region Methods for Multiscale Nonlinear Optimization
- Regularized Newton methods for minimizing functions with Hölder continuous hessians
- Simulated annealing algorithms for continuous global optimization: Convergence conditions
- Subsampled inexact Newton methods for minimizing large sums of convex functions
- Subsampled nonmonotone spectral gradient methods
- Testing Unconstrained Optimization Software
- The use of quadratic regularization with a cubic descent condition for unconstrained optimization
- Worst-case evaluation complexity for unconstrained nonlinear optimization using high-order regularized models
- Worst-case evaluation complexity of non-monotone gradient-related algorithms for unconstrained optimization
Cited in
(14)- On the worst-case evaluation complexity of non-monotone line search algorithms
- On the inexact scaled gradient projection method
- On the worst-case complexity of nonlinear stepsize control algorithms for convex unconstrained optimization
- Worst-case evaluation complexity of non-monotone gradient-related algorithms for unconstrained optimization
- Worst case complexity bounds for linesearch-type derivative-free algorithms
- Universal nonmonotone line search method for nonconvex multiobjective optimization problems with convex constraints
- A novel Riemannian conjugate gradient method with iteration complexity guarantees
- Subsampled cubic regularization method for finite-sum minimization
- A non-monotone line-search method for minimizing functions with spurious local minima
- A worst-case complexity analysis for Riemannian non-monotone line-search methods
- The generalized conditional gradient method for multiobjective composite optimization problems with non-monotone line search
- Iteration of complexity and asymptotic analysis of gradient-type methods with non-monotone line search on Riemannian manifolds
- Subsampled cubic regularization method with distinct sample sizes for function, gradient, and hessian
- Worst case complexity of direct search
This page was built for publication: A generalized worst-case complexity analysis for non-monotone line searches
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2028039)
