Improved bounds for sparse recovery from adaptive measurements

It is shown here that adaptivity in sampling results in dramatic improvements in the recovery of sparse signals in white Gaussian noise. An adaptive sampling-and-refinement procedure called distilled sensing is discussed and analyzed, resulting in fundamental new asymptotic scaling relationships in terms of the minimum feature strength required for reliable signal detection or localization (support recovery). In particular, reliable detection and localization using non-adaptive samples is possible only if the feature strength grows logarithmically in the problem dimension. Here it is shown that using adaptive sampling, reliable detection is possible provided the feature strength exceeds a constant, and localization is possible when the feature strength exceeds any (arbitrarily slowly) growing function of the problem dimension.

[1]  Robert D. Nowak,et al.  Distilled Sensing: Adaptive Sampling for Sparse Detection and Estimation , 2010, IEEE Transactions on Information Theory.

[2]  E.J. Candes.  Compressive Sampling , 2022 .

[3]  Robert D. Nowak,et al.  Compressive distilled sensing: Sparse recovery using adaptivity in compressive measurements , 2009, 2009 Conference Record of the Forty-Third Asilomar Conference on Signals, Systems and Computers.

[4]  Yu. I. Ingster,et al.  Nonparametric Goodness-of-Fit Testing Under Gaussian Models , 2002 .

[5]  David A. Cohn,et al.  Neural Network Exploration Using Optimal Experiment Design , 1993, NIPS.

[6]  Michael J. Swain,et al.  Promising directions in active vision , 1993, International Journal of Computer Vision.

[7]  D. Donoho,et al.  Higher criticism for detecting sparse heterogeneous mixtures , 2004, math/0410072.

[8]  Robert D. Nowak,et al.  Distilled sensing: selective sampling for sparse signal recovery , 2009, AISTATS.

[9]  L. Wasserman,et al.  Revisiting Marginal Regression , 2009, 0911.4080.

[10]  R. Elston,et al.  Optimal two‐stage genotyping in population‐based association studies , 2003, Genetic epidemiology.

[11]  I. Johnstone,et al.  Adapting to unknown sparsity by controlling the false discovery rate , 2005, math/0505374.

[12]  H. Schäfer,et al.  Including sampling and phenotyping costs into the optimization of two stage designs for genome wide association studies , 2007, Genetic epidemiology.

[13]  Peter Bauer,et al.  Two-stage designs for experiments with a large number of hypotheses , 2005, Bioinform..

[14]  D. Donoho,et al.  Asymptotic Minimaxity Of False Discovery Rate Thresholding For Sparse Exponential Data , 2006, math/0602311.

[15]  P. Bauer,et al.  Optimized multi‐stage designs controlling the false discovery or the family‐wise error rate , 2008, Statistics in medicine.

[16]  David A. Cohn,et al.  Active Learning with Statistical Models , 1996, NIPS.

[17]  J. Haupt,et al.  Adaptive discovery of sparse signals in noise , 2008, 2008 42nd Asilomar Conference on Signals, Systems and Computers.

[18]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[19]  Alfred O. Hero,et al.  Optimal Two-Stage Search for Sparse Targets Using Convex Criteria , 2008, IEEE Transactions on Signal Processing.