A Unifying Theory of Exactness of Linear Penalty Functions

Abstract

In this article we develop a theory of exact linear penalty functions that generalizes and unifies most of the results on exact penalization existing in the literature. We discuss several approaches to the study of both locally and globally exact linear penalty functions, and obtain various necessary and sufficient conditions for the exactness of a linear penalty function. We pay more attention than usual to necessary conditions that allows us to deeper understand the exact penalty technique.

arXiv commentThis is a slightly edited version of Accepted Manuscript of an article published by Taylor & Francis in Optimization on 22/12/2015, available online: http://www.tandfonline.com/10.1080/02331934.2015.1122005. In this version, a remark was added after Proposition 3.6 (the original numbering was preserved), and a mistake in the proof of Theorem 3.17 was corrected

References in corpus (1)

• Error Bounds and Metric Subregularity

Cited by in corpus (13)

• Exact Penalty Functions for Optimal Control Problems I: Main Theorem and Free-Endpoint Problems
• A unifying theory of exactness of linear penalty functions II: parametric penalty functions
• Existence of augmented Lagrange multipliers: reduction to exact penalty functions and localization principle
• Exact penalty functions with multidimensional penalty parameter and adaptive penalty updates
• Metric Regularity of Quasidifferentiable Mappings and Optimality Conditions for Nonsmooth Mathematical Programming Problems
• Exact Penalty Functions for Optimal Control Problems II: Exact Penalisation of Terminal and Pointwise State Constraints
• Augmented Lagrangian Functions for Cone Constrained Optimization: the Existence of Global Saddle Points and Exact Penalty Property
• A Unified Approach to the Global Exactness of Penalty and Augmented Lagrangian Functions I: Parametric Exactness
• New global optimality conditions for nonsmooth DC optimization problems
• A Unified Approach to the Global Exactness of Penalty and Augmented Lagrangian Functions II: Extended Exactness
• Steering exact penalty DCA for nonsmooth DC optimization problems with equality and inequality constraints
• Codifferentials and Quasidifferentials of the Expectation of Nonsmooth Random Integrands and Two-Stage Stochastic Programming
• Exact augmented Lagrangians for constrained optimization problems in Hilbert spaces I: Theory