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
  • [21] Topological Clustering via Adaptive Resonance Theory With Information Theoretic Learning
    Masuyama, Naoki
    Loo, Chu Kiong
    Ishibuchi, Hisao
    Kubota, Naoyuki
    Nojima, Yusuke
    Liu, Yiping
    IEEE ACCESS, 2019, 7 : 76920 - 76936
  • [22] Information Theoretic Criteria for Community Detection
    Branting, L. Karl
    ADVANCES IN SOCIAL NETWORK MINING AND ANALYSIS, 2010, 5498 : 114 - 130
  • [23] DETECTION OF SIGNALS BY INFORMATION THEORETIC CRITERIA
    WAX, M
    KAILATH, T
    IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1985, 33 (02): : 387 - 392
  • [24] EEG adaptive noise cancellation using information theoretic approach
    Darroudi, Ali
    Parchami, Jaber
    Razavi, Morteza Kafaee
    Sarbisheie, Ghazaleh
    BIO-MEDICAL MATERIALS AND ENGINEERING, 2017, 28 (04) : 325 - 338
  • [25] Feature extraction using information-theoretic learning
    Hild, Kenneth E., II
    Erdogmus, Deniz
    Torkkola, Kari
    Principe, Jose C.
    IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2006, 28 (09) : 1385 - 1392
  • [26] Prototype based classification using information theoretic learning
    Villmann, Th.
    Hammer, B.
    Schleif, F. -M.
    Geweniger, T.
    Fischer, T.
    Cottrell, M.
    NEURAL INFORMATION PROCESSING, PT 2, PROCEEDINGS, 2006, 4233 : 40 - 49
  • [27] InSAR noise reduction using adaptive dictionary learning
    Luo X.
    Suo Z.
    Liu Q.
    Xi'an Dianzi Keji Daxue Xuebao/Journal of Xidian University, 2016, 43 (01): : 18 - 23
  • [28] PET Image Deblurring Using Adaptive Dictionary Learning
    Valiollahzadeh, S.
    Clark, J.
    Mawlawi, O.
    MEDICAL PHYSICS, 2014, 41 (06)
  • [29] Design of low complexity multiuser detection using information theoretic criteria
    Wu, ML
    Fang, WHS
    Chen, JTS
    GLOBECOM '01: IEEE GLOBAL TELECOMMUNICATIONS CONFERENCE, VOLS 1-6, 2001, : 254 - 258
  • [30] Selection of a barley yield model using information-theoretic criteria
    Jasieniuk, Marie
    Taper, Mark L.
    Wagner, Nicole C.
    Stougaard, Robert N.
    Brelsford, Monica
    Maxwell, Bruce D.
    WEED SCIENCE, 2008, 56 (04) : 628 - 636