Adaptive-Size Dictionary Learning Using Information Theoretic Criteria

被引:6
|
作者
Dumitrescu, Bogdan [1 ]
Giurcaneanu, Ciprian Doru [2 ]
机构
[1] Univ Politehn Bucuresti, Dept Automat Control & Comp, 313 Spl Independentei, Bucharest 060042, Romania
[2] Univ Auckland, Dept Stat, Auckland 1142, New Zealand
关键词
dictionary learning; sparse representation; information theoretic criteria; dictionary size; SELECTION; DESIGN; SVD;
D O I
10.3390/a12090178
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Finding the size of the dictionary is an open issue in dictionary learning (DL). We propose an algorithm that adapts the size during the learning process by using Information Theoretic Criteria (ITC) specialized to the DL problem. The algorithm is built on top of Approximate K-SVD (AK-SVD) and periodically removes the less used atoms or adds new random atoms, based on ITC evaluations for a small number of candidate sub-dictionaries. Numerical experiments on synthetic data show that our algorithm not only finds the true size with very good accuracy, but is also able to improve the representation error in comparison with AK-SVD knowing the true size.
引用
收藏
页数:13
相关论文
共 50 条
  • [41] Using hierarchical information-theoretic criteria to optimize subsampling of extensive datasets
    Duarte, Belmiro P. M.
    Atkinson, Anthony C.
    Oliveira, Nuno M. C.
    CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS, 2024, 245
  • [42] SOURCE ENUMERATION USING THE PDF OF SAMPLE EIGENVALUES VIA INFORMATION THEORETIC CRITERIA
    Lu, Zhihua
    Zoubir, Abdelhak M.
    2012 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP), 2012, : 3361 - 3364
  • [43] Signal detection and estimation using atomic decomposition and information-theoretic criteria
    López-Risueño, G
    Grajal, J
    Yeste-Ojeda, OA
    2004 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL II, PROCEEDINGS: SENSOR ARRAY AND MULTICHANNEL SIGNAL PROCESSING SIGNAL PROCESSING THEORY AND METHODS, 2004, : 1097 - 1100
  • [44] Fuzzy classification using information theoretic learning vector quantization
    Villmann, Thomas
    Hammer, Barbara
    Schleif, Frank-Michael
    Hermann, Wieland
    Cottrell, Marie
    NEUROCOMPUTING, 2008, 71 (16-18) : 3070 - 3076
  • [45] Inferring connectivity of genetic regulatory networks using information-theoretic criteria
    Zhao, Wentao
    Serpedin, Erchin
    Dougherty, Edward R.
    IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS, 2008, 5 (02) : 262 - 274
  • [46] Classification of multiple sclerosis lesions using adaptive dictionary learning
    Deshpande, Hrishikesh
    Maurel, Pierre
    Barillot, Christian
    COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2015, 46 : 2 - 10
  • [47] A robust matching pursuit algorithm using information theoretic learning
    Zhang, Miaohua
    Gao, Yongsheng
    Sun, Changming
    Blumenstein, Michael
    PATTERN RECOGNITION, 2020, 107
  • [48] Information theoretic competitive learning in self-adaptive multi-layered networks
    Kamimura, R
    CONNECTION SCIENCE, 2003, 15 (01) : 3 - 26
  • [49] Estimation of the Channel-Impulse-Response Length for Adaptive OFDM Systems Based on Information Theoretic Criteria
    Ali, Ali Aassie
    Nguyen, Van Due
    Kyamakya, K.
    Omar, A. S.
    2006 IEEE 63RD VEHICULAR TECHNOLOGY CONFERENCE, VOLS 1-6, 2006, : 1888 - +
  • [50] Adaptive information agents using competitive learning
    Khan, I
    Card, HC
    JOURNAL OF NETWORK AND COMPUTER APPLICATIONS, 1998, 21 (02) : 69 - 89