A Generalized Tikhonov Regularization and A-Priori Choice of Regularization Parameters
Gongsheng Li
Abstract
Gongsheng Li
Abstract
With aid of the regularizing filters, a new class of regularization methods which called generalized Tikhonov regularization for solving the first kind equations with perturbed operators and noisy data is constructed. Applying singular systems of compact operators, the convergence and optimum asymptotic order of the regularized solution is obtained by a-priori choosing regularization parameters.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
With aid of the regularizing filters, a new class of regularization methods which called generalized Tikhonov regularization for solving the first kind equations with perturbed operators and noisy data is constructed. Applying singular systems of compact operators, the convergence and optimum asymptotic order of the regularized solution is obtained by a-priori choosing regularization parameters.
Key concepts: Tikhonov regularization, Backus–Gilbert method, Regularization perspectives on support vector machines, Regularization (linguistics), Mathematics, A priori and a posteriori, Applied mathematics, Proximal gradient methods for learning