Variational properties of value functions

TL;DR

This paper explores the variational properties of value functions in convex regularization problems, linking sensitivity to parameters with Lagrange multipliers, and introduces an inverse function theorem applicable beyond convex cases.

Contribution

It characterizes the variational behavior of value functions in convex and non-convex regularization, providing new theoretical tools for parameter selection and algorithm development.

Findings

Linked value function sensitivity to Lagrange multipliers.

Introduced an inverse function theorem for non-convex formulations.

Numerical examples validate theoretical insights.

Abstract

Regularization plays a key role in a variety of optimization formulations of inverse problems. A recurring theme in regularization approaches is the selection of regularization parameters, and their effect on the solution and on the optimal value of the optimization problem. The sensitivity of the value function to the regularization parameter can be linked directly to the Lagrange multipliers. This paper characterizes the variational properties of the value functions for a broad class of convex formulations, which are not all covered by standard Lagrange multiplier theory. An inverse function theorem is given that links the value functions of different regularization formulations (not necessarily convex). These results have implications for the selection of regularization parameters, and the development of specialized algorithms. Numerical examples illustrate the theoretical results.