Newton's Method for Large Bound-Constrained Optimization Problems
From MaRDI portal
The authors consider a trust region version of Newton's method for the optimization problem \(\min \{f(x): x \in \Omega\}\), \(\Omega=\{x \in R^n: l \leq x \leq u\}\). Given \(\eta_0>0\), the iterate \(x_k\) is updated by setting \(x_{k+1}=x_k +s_k\) if \(\rho_k=[f(x_k+s_k)-f(x_k)]/\psi_k (s_k) >\eta_0\) and \(x_{k+1}=x_k\) otherwise. The step \(s_k\) satisfies \(\psi(s_k) \leq \mu_0 \psi_k (s_k^C), \|s_k\|\leq \mu_1 \Delta_k, x_k +s_k \in \Omega\); \(\mu_0, \mu_1 >0\), \(\mu_0 <1/2\), where \(\psi_k (w)=\langle f'(x_k), w\rangle+ 1/2 \langle w, B_k w \rangle\) and \(B_k\) is a current approximation for the Hessian \(f(x_k)\). The algorithm is equipped with the rule for updating of the bound of the trust region \(\Delta_k\) at each iteration. The Cauchy step \(s_k^C=s_k (\alpha_k)\) is chosen subject to the constraints \(\psi_k (s_k (\alpha_k)) \leq \mu_0 \langle f'(x_k), s_k (\alpha_k)\rangle, \|s_k (\alpha_k)\|\leq \mu_1 \Delta_k\), where \(s_k (\alpha)= P[x_k-\alpha f'(x_k)]-x_k\) and \(P: R^n \to \Omega\) denotes the projection onto the feasible set \(\Omega\). A convergence theorem for uniformly bounded family \(\{ B_k\}\) and estimates for the rate of convergence for the case \(B_k=f(x_k)\) are established. The authors also present results of computational tests.
Recommendations
- Newton Methods For Large-Scale Linear Inequality-Constrained Minimization
- Newton Methods for Large-Scale Linear Equality-Constrained Minimization
- Newton's method for constrained optimization
- Newton's method for convex optimization
- Newton's method for singular constrained optimization problems
- scientific article; zbMATH DE number 2208632
- Newton-Type Methods for Optimization and Variational Problems
- scientific article; zbMATH DE number 4105720
- scientific article; zbMATH DE number 4024658
- scientific article; zbMATH DE number 5233132
Cited in
(only showing first 100 items - show all)- Adaptive discrete harmonic grid generation
- A sparse counterpart of Reichel and Gragg's package QRUP
- A survey of truncated-Newton methods
- A convergent least-squares regularized blind deconvolution approach
- TRON
- Local convergence analysis of projection-type algorithms: unified approach
- Globally and superlinearly convergent algorithms for the solution of box-constrained optimi\-zation
- An affine scaling trust-region approach to bound-constrained nonlinear systems
- Some recent advances in projection-type methods for variational inequalities
- An active set feasible method for large-scale minimization problems with bound constraints
- An active set modified Polak-Ribiére-Polyak method for large-scale nonlinear bound constrained optimization
- A hybrid differential dynamic programming algorithm for constrained optimal control problems. I: Theory
- Approximate solution of system of equations arising in interior-point methods for bound-constrained optimization
- Truncated trust region method for nonlinear inverse problems and application in full-waveform inversion
- A reduced proximal-point homotopy method for large-scale non-convex BQP
- A matrix-free trust-region Newton algorithm for convex-constrained optimization
- A primal-dual algorithm for risk minimization
- An active set quasi-Newton method with projection step for monotone nonlinear equations
- Limited memory BFGS algorithm for the matrix approximation problem in Frobenius norm
- A novel projected gradient-like method for optimization problems with simple constraints
- Scaled projected-directions methods with application to transmission tomography
- Non-negative moment fitting quadrature rules for fictitious domain methods
- Globally convergent DC trust-region methods
- A scalable algorithm for MAP estimators in Bayesian inverse problems with Besov priors
- Variational inequality approach to enforcing the non-negative constraint for advection-diffusion equations
- A relaxed nonmonotone adaptive trust region method for solving unconstrained optimization problems
- A matrix-free augmented Lagrangian algorithm with application to large-scale structural design optimization
- A two-stage active-set algorithm for bound-constrained optimization
- Interior-point solver for large-scale quadratic programming problems with bound constraints
- Superlinearly convergent trust-region method without the assumption of positive-definite Hessian
- An interior-point affine-scaling trust-region method for semismooth equations with box constraints
- Optimal weed population control using nonlinear programming
- Solving the maximum clique problem with symmetric rank-one non-negative matrix approximation
- An optimal subgradient algorithm for large-scale bound-constrained convex optimization
- An adaptive truncation criterion, for linesearch-based truncated Newton methods in large scale nonconvex optimization
- Inexact successive quadratic approximation for regularized optimization
- PAL-Hom method for QP and an application to LP
- An accurate active set conjugate gradient algorithm with project search for bound constrained optimization
- Harmonic hexahedral structured grid generation
- A limited memory quasi-Newton trust-region method for box constrained optimization
- An adaptive gradient algorithm for large-scale nonlinear bound constrained optimization
- A trust region method based on a new affine scaling technique for simple bounded optimization
- A non-monotonic method for large-scale non-negative least squares
- Newton-like methods for solving vector optimization problems
- Convergence of Affine-Scaling Interior-Point Methods with Line Search for Box Constrained Optimization
- Numerical optimization for constrained image registration
- An active-set trust-region method for derivative-free nonlinear bound-constrained optimization
- Newton Methods For Large-Scale Linear Inequality-Constrained Minimization
- An active set truncated Newton method for large-scale bound constrained optimization
- Modulus methods for nonnegatively constrained image restoration
- Implementing a smooth exact penalty function for equality-constrained nonlinear optimization
- An active set algorithm for nonlinear optimization with polyhedral constraints
- The lagrange-newton method for infinite-dimensional optimization problems
- Improving ultimate convergence of an augmented Lagrangian method
- A matrix-free approach to build band preconditioners for large-scale bound-constrained optimization
- On the convergence rate of scaled gradient projection method
- Modular proximal optimization for multidimensional total-variation regularization
- An adaptive nonmonotone trust region algorithm
- Optimization methods for large-scale machine learning
- Nonmonotone Barzilai-Borwein gradient algorithm for \(\ell_1\)-regularized nonsmooth minimization in compressive sensing
- Radius Margin Bounds for Support Vector Machines with the RBF Kernel
- A two-level decomposition framework exploiting first and second order information for SVM training problems
- Reduced approach for stochastic optimal control problems
- A framework for the upscaling of the electrical conductivity in the quasi-static Maxwell's equations
- scientific article; zbMATH DE number 7400716 (Why is no real title available?)
- PNKH-B: A Projected Newton--Krylov Method for Large-Scale Bound-Constrained Optimization
- A hybrid algorithm for solving minimization problem over (R,S)-symmetric matrices with the matrix inequality constraint
- A Smoothing Active Set Method for Linearly Constrained Non-Lipschitz Nonconvex Optimization
- Simultaneous sensing error recovery and tomographic inversion using an optimization-based approach
- The Conjugate Residual Method in Linesearch and Trust-Region Methods
- A reduced-space algorithm for minimizing _1-regularized convex functions
- Large-scale linear rankSVM
- The common-directions method for regularized empirical risk minimization
- A proximal subgradient projection algorithm for linearly constrained strictly convex problems
- A filter-trust-region method for simple-bound constrained optimization
- Newton method of solving Karush-Kuhn-Tucker systems for a constrained Minimax Problem
- Projected Gradient Methods for Nonnegative Matrix Factorization
- An accurate active set Newton algorithm for large scale bound constrained optimization.
- An alternating projected gradient algorithm for nonnegative matrix factorization
- Nonlinear optimization and support vector machines
- Nonlinear optimization and support vector machines
- iNEOS: An interactive environment for nonlinear optimization
- A relaxation-based probabilistic approach for PDE-constrained optimization under uncertainty with pointwise state constraints
- A proximal trust-region method for nonsmooth optimization with inexact function and gradient evaluations
- Local convergence analysis of an inexact trust-region method for nonsmooth optimization
- Modified active set projected spectral gradient method for bound constrained optimization
- An active set limited memory BFGS algorithm for bound constrained optimization
- Uniform flow in axisymmetric devices through permeability optimization
- Efficient proximal subproblem solvers for a nonsmooth trust-region method
- Image processing using Newton-based algorithm of nonnegative matrix factorization
- Scalable optimal control for inequality-constrained discretizations of scalar conservation laws
- Leveraging GPU batching for scalable nonlinear programming through massive Lagrangian decomposition
- Numerical methods for A-optimal designs with a sparsity constraint for ill-posed inverse problems
- An augmented Lagrangian trust-region method with inexact gradient evaluations to accelerate constrained optimization problems using model hyperreduction
- Gradient projection method for enforcing crack irreversibility as box constraints in a robust monolithic phase-field scheme
- Sample size selection in optimization methods for machine learning
- An augmented Lagrangian algorithm for constrained nonlinear least-squares
- Memory-efficient nonsmooth dynamic optimization using adaptive randomized compression
- A convergent decomposition method for box-constrained optimization problems
- Least squares problems with inequality constraints as quadratic constraints
This page was built for publication: Newton's Method for Large Bound-Constrained Optimization Problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4702324)
