[工程科技]A forward-backward splitting algorithm for the minimization of non-smooth convex functional.pdfVIP

A forward-backward splitting algorithm for the minimization of non-smooth convex functionals in Banach space 8 Kristian Bredies 0 0 2 July 4, 2008 l u J Abstract 4 We consider the task of computing an approximate minimizer of ] the sum of a smooth and non-smooth convex functional, respectively, A in Banach space. Motivated by the classical forward-backward split- N ting method for the subgradients in Hilbert space, we propose a gen- . h eralization which involves the iterative solution of simpler subprob- t lems. Descent and convergence properties of this new algorithm are a m studied. Furthermore, the results are applied to the minimization of Tikhonov-functionals associated with linear inverse problems and [ semi-norm penalization in Banach spaces. With the help of Bregman- 1 Taylor-distance estimates, rates of convergence for the forward-back- v ward splitting procedure are obtained. Examples which demonstrate 8 the applicability are given, in particular, a generalization of the iter- 7 7 ative soft-thresholding method by Daubechies, Defrise and De Mol to 0 Banach spaces is presented. . 7 0 Keywords: forward-backward splitting, convergence analysis, Banach 8 spaces, inverse problems, iterative thresholding