Adaptive cubic regularization methods with dynamic inexact Hessian information and applications to finite-sum minimization
From MaRDI portal
(Redirected from Publication:4964100)
Abstract: We consider the Adaptive Regularization with Cubics approach for solving nonconvex optimization problems and propose a new variant based on inexact Hessian information chosen dynamically. The theoretical analysis of the proposed procedure is given. The key property of ARC framework, constituted by optimal worst-case function/derivative evaluation bounds for first- and second-order critical point, is guaranteed. Application to large-scale finite-sum minimization based on subsampled Hessian is discussed and analyzed in both a deterministic and probabilistic manner and equipped with numerical experiments on synthetic and real datasets.
Recommendations
- Stochastic analysis of an adaptive cubic regularization method under inexact gradient evaluations and dynamic Hessian accuracy
- Newton-type methods for non-convex optimization under inexact Hessian information
- A cubic regularization of Newton's method with finite difference Hessian approximations
- 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
Cited in
(29)- An algorithm for the minimization of nonsmooth nonconvex functions using inexact evaluations and its worst-case complexity
- A generalized worst-case complexity analysis for non-monotone line searches
- Adaptive regularization for nonconvex optimization using inexact function values and randomly perturbed derivatives
- Linesearch Newton-CG methods for convex optimization with noise
- A stochastic first-order trust-region method with inexact restoration for finite-sum minimization
- A cubic regularization of Newton's method with finite difference Hessian approximations
- Subsampled nonmonotone spectral gradient methods
- Inexact restoration with subsampled trust-region methods for finite-sum minimization
- Combining stochastic adaptive cubic regularization with negative curvature for nonconvex optimization
- A sequential adaptive regularisation using cubics algorithm for solving nonlinear equality constrained optimization
- The impact of noise on evaluation complexity: the deterministic trust-region case
- Stochastic analysis of an adaptive cubic regularization method under inexact gradient evaluations and dynamic Hessian accuracy
- Convergence of Newton-MR under inexact Hessian information
- Stochastic variance-reduced cubic regularization methods
- Adaptive Regularization Algorithms with Inexact Evaluations for Nonconvex Optimization
- LSOS: Line-search second-order stochastic optimization methods for nonconvex finite sums
- A filter sequential adaptive cubic regularization algorithm for nonlinear constrained optimization
- Derivative-free separable quadratic modeling and cubic regularization for unconstrained optimization
- Inexact tensor methods and their application to stochastic convex optimization
- Subsampled first-order optimization methods with applications in imaging
- Complexity guarantees for nonconvex Newton-MR under inexact Hessian information
- Inexact Gauss-Newton methods with matrix approximation by sampling for nonlinear least-squares and systems
- Yet another fast variant of Newton's method for nonconvex optimization
- A stochastic objective-function-free adaptive regularization method with optimal complexity
- Subsampled cubic regularization method for finite-sum minimization
- A line-search filter adaptive regularization algorithm using cubics based on affine scaling interior-point methods for constrained optimization
- An adaptive cubic regularisation algorithm based on interior-point methods for optimization with general inequality constraints
- Subsampled cubic regularization method with distinct sample sizes for function, gradient, and hessian
- Escaping saddle points in distributed nonconvex optimization via cubic regularization
This page was built for publication: Adaptive cubic regularization methods with dynamic inexact Hessian information and applications to finite-sum minimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4964100)
