Lipschitz behavior of the robust regularization

DOI10.1137/08073682XMaRDI QIDQ3162578zbMATH OpenFDO

Authors A. S. Lewis, C. H. Jeffrey Pang

Publication date 20 October 2010

Published in SIAM Journal on Control and Optimization (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0810.0098

zbMATH Keywords

regularizationnonsmooth analysisrobust controlrobust optimizationlocally Lipschitzsemi-algebraic functionssemidefinite representable regularization

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Eigenvalues, singular values, and eigenvectors (15A18) Nonlinear programming (90C30) Sensitivity (robustness) (93B35) Regularity of solutions in optimal control (49N60) Semialgebraic sets and related spaces (14P10)

Abstract: To minimize or upper-bound the value of a function "robustly", we might instead minimize or upper-bound the "epsilon-robust regularization", defined as the map from a point to the maximum value of the function within an epsilon-radius. This regularization may be easy to compute: convex quadratics lead to semidefinite-representable regularizations, for example, and the spectral radius of a matrix leads to pseudospectral computations. For favorable classes of functions, we show that the robust regularization is Lipschitz around any given point, for all small epsilon > 0, even if the original function is nonlipschitz (like the spectral radius). One such favorable class consists of the semi-algebraic functions. Such functions have graphs that are finite unions of sets defined by finitely-many polynomial inequalities, and are commonly encountered in applications.

Recommendations

• Regularization by erasement
• Approximation and regularization of Lipschitz functions: Convergence of the gradients
• Robust analysis and global optimization
• scientific article; zbMATH DE number 653049
• On robustness of the regularity property of maps

Cited in

(14)

• A unified approach through image space analysis to robustness in uncertain optimization problems
• A look at robustness and stability of \(\ell_1\)-versus \(\ell_0\)-regularization: discussion of papers by Bertsimas et al. and Hastie et al.
• Fabrication-adaptive optimization with an application to photonic crystal design
• Characterization of the Clarke regularity of subanalytic sets
• Decision uncertainty in multiobjective optimization
• Characterizations for optimality conditions of general robust optimization problems
• A unified characterization of multiobjective robustness via separation
• Characterization of the equivalence of robustification and regularization in linear and matrix regression
• Diametrical risk minimization: theory and computations
• Distributionally robust optimization. A review on theory and applications
• An algorithmic approach to multiobjective optimization with decision uncertainty
• \(l_{p}\)-norm regularization method (\( 0<p<1 \)) and DC programming for correction system of inconsistency linear inequalities
• Smoothing methods for nonsmooth, nonconvex minimization
• The point-based robustness gap for uncertain multiobjective optimization

Describes a project that uses

Uses Software

• Eigtool

This page was built for publication: Lipschitz behavior of the robust regularization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3162578)