A unified approach to convergence rates for ℓ1-regularization and lacking sparsity

Abstract In ℓ1-regularization, which is an important tool in signal and image processing, one usually is concerned with signals and images having a sparse representation in some suitable basis, e.g., in a wavelet basis. Many results on convergence and convergence rates of sparse approximate solutions to linear ill-posed problems are known, but rate results for the ℓ1-regularization in case of lacking sparsity had not been published until 2013. In the last two years, however, two articles appeared providing sufficient conditions for convergence rates in case of non-sparse but almost sparse solutions. In the present paper, we suggest a third sufficient condition, which unifies the existing two and, by the way, also incorporates the well-known restricted isometry property.

[1]  Bernd Hofmann,et al.  Regularization properties of the sequential discrepancy principle for Tikhonov regularization in Banach spaces , 2014 .

[2]  Bernd Hofmann,et al.  Parameter choice in Banach space regularization under variational inequalities , 2012 .

[3]  Otmar Scherzer,et al.  Variational Methods in Imaging , 2008, Applied mathematical sciences.

[4]  O. Scherzer,et al.  A convergence rates result for Tikhonov regularization in Banach spaces with non-smooth operators , 2007 .

[5]  O. Scherzer,et al.  Generalized Convergence Rates Results for Linear Inverse Problems in Hilbert Spaces , 2014, 1409.7610.

[6]  Convergence rates in ℓ1-regularization when the basis is not smooth enough , 2013, 1311.1923.

[7]  D. Lorenz,et al.  Convergence rates and source conditions for Tikhonov regularization with sparsity constraints , 2008, 0801.1774.

[8]  E. Candès,et al.  Stable signal recovery from incomplete and inaccurate measurements , 2005, math/0503066.

[9]  Bernd Hofmann,et al.  An extension of the variational inequality approach for obtaining convergence rates in regularization of nonlinear ill-posed problems , 2010 .

[10]  M. Grasmair.  Generalized Bregman distances and convergence rates for non-convex regularization methods , 2010 .

[11]  Martin Burger,et al.  ERROR ESTIMATES FOR GENERAL FIDELITIES , 2011 .

[12]  B. Hofmann,et al.  The impact of a curious type of smoothness conditions on convergence rates in l1-regularization , 2013 .

[13]  M. Z. Nashed,et al.  A NEW APPROACH TO CLASSIFICATION AND REGULARIZATION OF ILL-POSED OPERATOR EQUATIONS , 1987 .

[14]  Ronny Ramlau,et al.  CONVERGENCE RATES FOR REGULARIZATION WITH SPARSITY CONSTRAINTS , 2010 .

[15]  B. Hofmann,et al.  On the interplay of basis smoothness and specific range conditions occurring in sparsity regularization , 2013, 1307.7493.

[16]  K. Bredies,et al.  Regularization with non-convex separable constraints , 2009 .

[17]  Thorsten Hohage,et al.  Convergence rates in expectation for Tikhonov-type regularization of inverse problems with Poisson data , 2012, 1204.1669.

[18]  Martin Burger,et al.  Convergence rates in $\mathbf{\ell^1}$-regularization if the sparsity assumption fails , 2012 .

[19]  Jens Flemming,et al.  On ℓ1-Regularization in Light of Nashed's Ill-Posedness Concept , 2015, Comput. Methods Appl. Math..