Sparse Bayesian learning for basis selection

被引:1123
作者
Wipf, DP [1 ]
Rao, BD [1 ]
机构
[1] Univ Calif San Diego, Dept Elect & Comp Engn, La Jolla, CA 92092 USA
关键词
basis selection; diversity measures; linear inverse problems; sparse Bayesian learning; sparse representations;
D O I
10.1109/TSP.2004.831016
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Sparse Bayesian learning (SBL) and specifically relevance vector machines have received much attention in the machine learning literature as a means of achieving parsimonious representations in the context of regression and classification. The methodology relies on a parameterized prior that encourages models with few nonzero weights. In this paper, we adapt SBL to the signal processing problem of basis selection from overcomplete dictionaries, proving, several results About the SBL cost function that elucidate its general behavior and provide solid theoretical justification for this application. Specifically, we have shown that SBL retains a desirable property of the l(0)-norm diversity measure (i.e., the global minimum is achieved at the maximally sparse solution) while often possessing a more limited constellation of local minima. We have also demonstrated that the local minima that do exist are achieved at sparse solutions. Later, we provide a novel interpretation of SBL that gives us valuable insight into why it is successful in producing sparse representations. Finally, we include simulation studies comparing sparse Bayesian learning with Basis Pursuit and the more recent FOCal Underdetermined System Solver (FOCUSS) class of basis selection algorithms. These results indicate that our theoretical insights translate directly into improved performance.
引用
收藏
页码:2153 / 2164
页数:12
相关论文
共 36 条
[1]  
BARRODALE I, 1970, P S NONLINEAR PROGRA, P447
[2]  
BERGER J. O., 2013, Statistical Decision Theory and Bayesian Analysis, DOI [10.1007/978-1-4757-4286-2, DOI 10.1007/978-1-4757-4286-2]
[3]  
Bishop C. M., 2000, P 16 C UNC ART INT S, P46
[4]   EXTRAPOLATION AND SPECTRAL ESTIMATION WITH ITERATIVE WEIGHTED NORM MODIFICATION [J].
CABRERA, SD ;
PARKS, TW .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1991, 39 (04) :842-851
[5]   SPARSE APPROXIMATE MULTIQUADRIC INTERPOLATION [J].
CARLSON, RE ;
NATARAJAN, BK .
COMPUTERS & MATHEMATICS WITH APPLICATIONS, 1994, 27 (06) :99-108
[6]   FAST ORTHOGONAL LEAST-SQUARES ALGORITHM FOR EFFICIENT SUBSET MODEL SELECTION [J].
CHEN, S ;
WIGGER, J .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 1995, 43 (07) :1713-1715
[7]   Atomic decomposition by basis pursuit [J].
Chen, SSB ;
Donoho, DL ;
Saunders, MA .
SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1998, 20 (01) :33-61
[8]   Sparse channel estimation via matching pursuit with application to equalization [J].
Cotter, SF ;
Rao, BD .
IEEE TRANSACTIONS ON COMMUNICATIONS, 2002, 50 (03) :374-377
[9]   Optimally sparse representation in general (nonorthogonal) dictionaries via l1 minimization [J].
Donoho, DL ;
Elad, M .
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2003, 100 (05) :2197-2202
[10]   Uncertainty principles and ideal atomic decomposition [J].
Donoho, DL ;
Huo, XM .
IEEE TRANSACTIONS ON INFORMATION THEORY, 2001, 47 (07) :2845-2862