Quadratic Optimization in Ill-Posed Problems
نویسندگان
چکیده
Ill posed quadratic optimization frequently occurs in control and inverse problems and are not covered by the Lax-Milgram-Riesz theory. Typically small changes in the input data can produce very large oscillations on the output. We investigate the conditions under which the minimum value of the cost function is finite and we explore the ‘hidden connection’ between the optimization problem and the least-squares method. Eventually, we address some examples coming from optimal control and data completion, showing how relevant our contribution is in the knowledge of what happens for various ill-posed problems. The results we state bring a substantial improvement to the analysis of the regularization methods applied to the ill-posed quadratic optimization problems. keywords: Quadratic optimization, least-squares, ill-posedness, Picard’s principle, optimal control, data completion.
منابع مشابه
Ill-Posed and Linear Inverse Problems
In this paper ill-posed linear inverse problems that arises in many applications is considered. The instability of special kind of these problems and it's relation to the kernel, is described. For finding a stable solution to these problems we need some kind of regularization that is presented. The results have been applied for a singular equation.
متن کاملActive-Set Methods for Support Vector Machines
This chapter describes an active-set algorithm for the solution of quadratic programming problems in the context of Support Vector Machines (SVMs). Most of the common SVM optimizers implement working-set algorithms like the SMO method because of their ability to handle large data sets. Although they show generally good results, they may perform weakly in some situations, e.g., if the problem is...
متن کاملRegularization of autoconvolution and other ill-posed quadratic equations by decomposition
Standard methods for regularizing ill-posed nonlinear equations rely on derivatives of the nonlinear forward mapping. Thereby stronger structural properties of the concrete problem are neglected and the derived algorithms only show mediocre efficiency. We concentrate on nonlinear mappings with quadratic structure and develop a derivative-free regularization method that allows us to apply classi...
متن کاملNumerical methods for optimal experimental design of large-scale ill-posed problems
Experimental design for over-determined problems is a well studied topic where different criteria and optimization algorithms are explored. For ill-posed problems experimental design is rather new. In this paper we discuss optimal experimental design for ill-posed problems and suggest a numerical framework to efficiently achieve such a design. We demonstrate the effectiveness of our algorithm a...
متن کاملLAP: a Linearize and Project Method for Solving Inverse Problems with Coupled Variables
Many inverse problems involve two or more sets of variables that represent different physical quantities but are tightly coupled with each other. For example, image super-resolution requires joint estimation of image and motion parameters from noisy measurements. Exploiting this structure is key for efficiently solving large-scale problems to avoid, e.g., ill-conditioned optimization problems. ...
متن کامل