Sequential Testing for Sparse Recovery

被引:29
作者
Malloy, Matthew L. [1 ]
Nowak, Robert D. [1 ]
机构
[1] Univ Wisconsin, Dept Elect & Comp Engn, Madison, WI 53706 USA
关键词
Sequential analysis; sparse recovery; SPRT; sequential thresholding; multi-armed bandits; 2-STAGE DESIGNS; SEARCH;
D O I
10.1109/TIT.2014.2363846
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper studies sequential methods for recovery of sparse signals in high dimensions. When compared with fixed sample size procedures, in the sparse setting, sequential methods can result in a large reduction in the number of samples needed for reliable signal support recovery. Starting with a lower bound, we show any coordinate-wise sequential sampling procedure fails in the high dimensional limit provided the average number of measurements per dimension is less then log(s)/D(P-0 parallel to P-1), where s is the level of sparsity and D(P-0 parallel to P-1) is the Kullback-Leibler divergence between the underlying distributions. A series of sequential probability ratio tests, which require complete knowledge of the underlying distributions is shown to achieve this bound. Motivated by real-world experiments and recent work in adaptive sensing, we introduce a simple procedure termed sequential thresholding, which can be implemented when the underlying testing problem satisfies a monotone likelihood ratio assumption. Sequential thresholding guarantees exact support recovery provided the average number of measurements per dimension grows faster than log(s)/D(P-0 parallel to P-1), achieving the lower bound. For comparison, we show any nonsequential procedure fails provided the number of measurements grows at a rate less than log(n)/D(P-1 parallel to|P-0), where n is the total dimension of the problem.
引用
收藏
页码:7862 / 7873
页数:12
相关论文
共 35 条
[1]   A MODIFICATION OF THE SEQUENTIAL PROBABILITY RATIO TEST TO REDUCE THE SAMPLE-SIZE [J].
ANDERSON, TW .
ANNALS OF MATHEMATICAL STATISTICS, 1960, 31 (01) :165-197
[2]  
[Anonymous], 2005, ELEMENTS INFORM THEO, DOI DOI 10.1002/047174882X
[3]   Optimal Two-Stage Search for Sparse Targets Using Convex Criteria [J].
Bashan, Eran ;
Raich, Raviv ;
Hero, Alfred O. .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2008, 56 (11) :5389-5402
[4]   Two-Stage Multiscale Search for Sparse Targets [J].
Bashan, Eran ;
Newstadt, Gregory ;
Hero, Alfred O., III .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2011, 59 (05) :2331-2341
[5]  
Bubeck S, 2009, LECT NOTES ARTIF INT, V5809, P23, DOI 10.1007/978-3-642-04414-4_7
[6]  
CASTRO R.M., 2012, Adaptive Sensing Performance Lower Bounds for Sparse Signal Estimation and Testing
[7]  
Even-Dar E., 2002, Computational Learning Theory. 15th Annual Conference on Computational Learning Theory, COLT 2002. Proceedings (Lecture Notes in Artificial Intelligence Vol.2375), P255
[8]  
Ghosh B., 1991, HDB SEQUENTIAL ANAL
[9]  
Ghosh BK., 1970, SEQUENTIAL TESTS STA
[10]  
GHOSH MN, 1960, J ROY STAT SOC B, V22, P360