Similarity Learning for High-Dimensional Sparse Data

被引:0
|
作者
Liu, Kuan [1 ]
Bellet, Aurelien [2 ]
Sha, Fei [1 ]
机构
[1] Univ Southern Calif, Los Angeles, CA 90007 USA
[2] Telecom ParisTech, Paris, France
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A good measure of similarity between data points is crucial to many tasks in machine learning. Similarity and metric learning methods learn such measures automatically from data, but they do not scale well respect to the dimensionality of the data. In this paper, we propose a method that can learn efficiently similarity measure from high-dimensional sparse data. The core idea is to parameterize the similarity measure as a convex combination of rank-one matrices with specific sparsity structures. The parameters are then optimized with an approximate Frank-Wolfe procedure to maximally satisfy relative similarity constraints on the training data. Our algorithm greedily incorporates one pair of features at a time into the similarity measure, providing an efficient way to control the number of active features and thus reduce overfitting. It enjoys very appealing convergence guarantees and its time and memory complexity depends on the sparsity of the data instead of the dimension of the feature space. Our experiments on real-world high-dimensional datasets demonstrate its potential for classification, dimensionality reduction and data exploration.
引用
收藏
页码:653 / 662
页数:10
相关论文
共 50 条
  • [41] Sparse Bayesian variable selection for classifying high-dimensional data
    Yang, Aijun
    Lian, Heng
    Jiang, Xuejun
    Liu, Pengfei
    [J]. STATISTICS AND ITS INTERFACE, 2018, 11 (02) : 385 - 395
  • [42] Sparse kernel k-means for high-dimensional data
    Guan, Xin
    Terada, Yoshikazu
    [J]. PATTERN RECOGNITION, 2023, 144
  • [43] Sparse redundancy analysis of high-dimensional genetic and genomic data
    Csala, Attila
    Voorbraak, Frans P. J. M.
    Zwinderman, Aeilko H.
    Hof, Michel H.
    [J]. BIOINFORMATICS, 2017, 33 (20) : 3228 - 3234
  • [44] Quantization techniques for similarity search in high-dimensional data spaces
    Garcia-Arellano, C
    Sevcik, K
    [J]. NEW HORIZONS IN INFORMATION MANAGEMENT, 2003, 2712 : 75 - 94
  • [45] Learning Sparse High-Dimensional Matrix-Valued Graphical Models From Dependent Data
    Tugnait, Jitendra K.
    [J]. IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2024, 72 : 3363 - 3379
  • [46] ON THE PERFORMANCE OF KERNEL ESTIMATORS FOR HIGH-DIMENSIONAL, SPARSE BINARY DATA
    GRUND, B
    HALL, P
    [J]. JOURNAL OF MULTIVARIATE ANALYSIS, 1993, 44 (02) : 321 - 344
  • [47] Fused Feature Representation Discovery for High-Dimensional and Sparse Data
    Suzuki, Jun
    Nagata, Masaaki
    [J]. PROCEEDINGS OF THE TWENTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, 2014, : 1593 - 1599
  • [48] Sparse representation approaches for the classification of high-dimensional biological data
    Li, Yifeng
    Ngom, Alioune
    [J]. BMC SYSTEMS BIOLOGY, 2013, 7
  • [49] CLASSIFICATION OF HIGH-DIMENSIONAL DATA USING THE SPARSE MATRIX TRANSFORM
    Bachega, Leonardo R. |
    Bouman, Charles A.
    [J]. 2010 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, 2010, : 265 - 268
  • [50] Sparse boosting for high-dimensional survival data with varying coefficients
    Yue, Mu
    Li, Jialiang
    Ma, Shuangge
    [J]. STATISTICS IN MEDICINE, 2018, 37 (05) : 789 - 800