Title :
Thresholds for the Recovery of Sparse Solutions via L1 Minimization
Author :
Donoho, David L. ; Tanner, Jared
Author_Institution :
Dept. of Stat., Stanford Univ., CA
Abstract :
The ubiquitous least squares method for systems of linear equations returns solutions which typically have all non-zero entries. However, solutions with the least number of non-zeros allow for greater insight. An exhaustive search for the sparsest solution is intractable, NP-hard. Recently, a great deal of research showed that linear programming can find the sparsest solution for certain ´typical´ systems of equations, provided the solution is sufficiently sparse. In this note we report recent progress determining conditions under which the sparsest solution to large systems is available by linear programming. Our work shows that there are sharp thresholds on sparsity below which these methods will succeed and above which they fail; it evaluates those thresholds precisely and hints at several interesting applications.
Keywords :
least squares approximations; linear programming; sparse matrices; L1 minimization; NP-hard; linear programming; sparse solution; ubiquitous least squares method; Cities and towns; Compressed sensing; Equations; Least squares methods; Linear programming; Mathematics; Minimization methods; Sampling methods; Sparse matrices; Statistics;
Conference_Titel :
Information Sciences and Systems, 2006 40th Annual Conference on
Conference_Location :
Princeton, NJ
Print_ISBN :
1-4244-0349-9
Electronic_ISBN :
1-4244-0350-2
DOI :
10.1109/CISS.2006.286462