Harnessing Structure in Composite Nonsmooth Minimization

DOI10.1137/22M1505827arXiv2206.15053OpenAlexW4284693804MaRDI QIDQ6046827FDOQ6046827

Authors: Gilles Bareilles, Franck Iutzeler, Jérôme Malick 

Publication date: 6 September 2023

Published in: SIAM Journal on Optimization (Search for Journal in Brave)

Abstract: We consider the problem of minimizing the composition of a nonsmooth function with a smooth mapping in the case where the proximity operator of the nonsmooth function can be explicitly computed. We first show that this proximity operator can provide the exact smooth substructure of minimizers, not only of the nonsmooth function, but also of the full composite function. We then exploit this proximal identification by proposing an algorithm which combines proximal steps with sequential quadratic programming steps. We show that our method locally identifies the optimal smooth substructure and then converges quadratically. We illustrate its behavior on two problems: the minimization of a maximum of quadratic functions and the minimization of the maximal eigenvalue of a parametrized matrix.

Full work available at URL: https://arxiv.org/abs/2206.15053

zbMATH Keywords

sequential quadratic programmingnonsmooth optimizationproximal operatorpartial smoothnessmanifold identificationmaximum eigenvalue minimization

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Methods of successive quadratic programming type (90C55) Nonconvex programming, global optimization (90C26) Sensitivity analysis for optimization problems on manifolds (49Q12)

Cites Work

• Numerical optimization. Theoretical and practical aspects. Transl. from the French
• Julia: a fresh approach to numerical computing
• Title not available (Why is that?)
• Composite proximal bundle method
• A proximal method for composite minimization
• Title not available (Why is that?)
• Nonsmooth optimization via quasi-Newton methods
• Computing proximal points of nonconvex functions
• A \(\mathcal{VU}\)-algorithm for convex minimization
• Newton methods for nonsmooth convex minimization: connections among \(\mathcal U\)-Lagrangian, Riemannian Newton and SQP methods
• Geometrical interpretation of the predictor-corrector type algorithms in structured optimization problems
• Title not available (Why is that?)
• Active Sets, Nonsmoothness, and Sensitivity
• On a Class of Nonsmooth Composite Functions
• Spectral bundle methods for non-convex maximum eigenvalue functions: second-order methods
• Partial Smoothness, Tilt Stability, and Generalized Hessians
• On Eigenvalue Optimization
• The $\U$-Lagrangian of the Maximum Eigenvalue Function
• An algorithm for composite nonsmooth optimization problems
• The multiproximal linearization method for convex composite problems
• Majorization-minimization procedures and convergence of SQP methods for semi-algebraic and tame programs
• The spectral bundle method with second-order information
• Nonsmooth optimization using Taylor-like models: error bounds, convergence, and termination criteria
• Newton acceleration on manifolds identified by proximal gradient methods
• An Introduction to Optimization on Smooth Manifolds
• Survey Descent: A Multipoint Generalization of Gradient Descent for Nonsmooth Optimization

This page was built for publication: Harnessing Structure in Composite Nonsmooth Minimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6046827)