A Linearly Convergent Algorithm for Solving a Class of Nonconvex/Affine Feasibility Problems
We introduce a class of nonconvex/affine feasibility problems (NCF), that consists of finding a point in the intersection of affine constraints with a nonconvex closed set. This class captures some interesting fundamental and NP hard problems arising in various application areas such as sparse recovery of signals and affine rank minimization that we briefly review. Exploiting the special structure of (NCF), we present a simple gradient projection scheme which is proven to converge to a unique solution of (NCF) at a linear rate under a natural assumption explicitly given defined in terms of the problem’s data.
KeywordsNonconvex affine feasibility Inverse problems Gradient projection algorithm Linear rate of convergence Scalable restricted isometry Mutual coherence of a matrix Sparse signal recovery Compressive sensing Affine rank minimization
Unable to display preview. Download preview PDF.
We thank two anonymous referees for their useful comments and suggestions. This research was partially supported by the Israel Science Foundation under ISF Grant 489-06.
- 2.Beck, A., Teboulle., M.: A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J. Imaging Sciences 2, 183–202 (2009)Google Scholar
- 3.Beck, A., Teboulle, M.: Gradient-based algorithms with applications to signal recovery problems. In: D. Palomar, Y. Eldar (eds.) Convex Optimization in Signal Processing and Communications, 42–88. Cambridge University Press (2010)Google Scholar
- 4.Bertsekas, D.: Non-Linear Programming, 2nd ed. Athena Scientific, Belmont, MA (1999)Google Scholar
- 7.Cadzow, J.A.: Signal enhancement – a composite property mapping algorithm. IEEE Trans. Acoustics, Speech, Signal Process. 36, 49–62 (1988)Google Scholar
- 17.Engl, H.W., Hanke, M., Neubauer, A.: Regularization of inverse problems, volume 375 of Mathematics and its Applications. Kluwer Academic Publishers Group, Dordrecht (1996)Google Scholar
- 18.Fazel, M., Hindi, H., Boyd, S.: Rank minimization and applications in system theory. In: American Control Conference, 3272–3278 (2004)Google Scholar
- 19.Golub, G., Loan, C.V.: Matrix computations, 3rd edn. Johns Hopkins University Press (1996)Google Scholar