On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers

被引:0
|
作者
Tapio Pahikkala
Antti Airola
Fabian Gieseke
Oliver Kramer
机构
[1] Department of Information Technology,
[2] University of Turku,undefined
[3] Department of Computer Science,undefined
[4] University of Copenhagen,undefined
[5] Computer Science Department,undefined
[6] Carl von Ossietzky University of Oldenburg,undefined
关键词
unsupervised learning; multi-class regularized least-squares classification; maximum margin clustering; combinatorial optimization;
D O I
暂无
中图分类号
学科分类号
摘要
In this work we present the first efficient algorithm for unsupervised training of multi-class regularized least-squares classifiers. The approach is closely related to the unsupervised extension of the support vector machine classifier known as maximum margin clustering, which recently has received considerable attention, though mostly considering the binary classification case. We present a combinatorial search scheme that combines steepest descent strategies with powerful meta-heuristics for avoiding bad local optima. The regularized least-squares based formulation of the problem allows us to use matrix algebraic optimization enabling constant time checks for the intermediate candidate solutions during the search. Our experimental evaluation indicates the potential of the novel method and demonstrates its superior clustering performance over a variety of competing methods on real world datasets. Both time complexity analysis and experimental comparisons show that the method can scale well to practical sized problems.
引用
收藏
页码:90 / 104
页数:14
相关论文
共 50 条
  • [1] On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers
    Tapio Pahikkala
    Antti Airola
    Fabian Gieseke
    Oliver Kramer
    [J]. Journal of Computer Science & Technology, 2014, 29 (01) : 90 - 104
  • [2] On Unsupervised Training of Multi-Class Regularized Least-Squares Classifiers
    Pahikkala, Tapio
    Airola, Antti
    Gieseke, Fabian
    Kramer, Oliver
    [J]. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY, 2014, 29 (01) : 90 - 104
  • [3] Unsupervised Multi-Class Regularized Least-Squares Classification
    Pahikkala, Tapio
    Airola, Antti
    Gieseke, Fabian
    Kramer, Oliver
    [J]. 12TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2012), 2012, : 585 - 594
  • [4] A Sparse Multi-class Least-Squares Support Vector Machine
    Xia, Xiao-Lei
    Li, Kang
    [J]. 2008 IEEE INTERNATIONAL SYMPOSIUM ON INDUSTRIAL ELECTRONICS, VOLS 1-5, 2008, : 648 - 653
  • [5] Superfast-Trainable Multi-Class Probabilistic Classifier by Least-Squares Posterior Fitting
    Sugiyama, Masashi
    [J]. IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2010, E93D (10) : 2690 - 2701
  • [6] RLScore: Regularized Least-Squares Learners
    Pahikkala, Tapio
    Airola, Antti
    [J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2016, 17
  • [7] CONSTRAINED REGULARIZED LEAST-SQUARES PROBLEMS
    ELFVING, T
    [J]. LECTURE NOTES IN MATHEMATICS, 1991, 1497 : 153 - 166
  • [8] Regularized Least-Squares for parse ranking
    Tsivtsivadze, E
    Pahikkala, T
    Pyysalo, S
    Boberg, J
    Mylläri, A
    Salakoski, T
    [J]. ADVANCES IN INTELLIGENT DATA ANALYSIS VI, PROCEEDINGS, 2005, 3646 : 464 - 474
  • [9] Discriminatively regularized least-squares classification
    Xue, Hui
    Chen, Songcan
    Yang, Qiang
    [J]. PATTERN RECOGNITION, 2009, 42 (01) : 93 - 104
  • [10] Optimal rates for the regularized least-squares algorithm
    Caponnetto, A.
    De Vito, E.
    [J]. FOUNDATIONS OF COMPUTATIONAL MATHEMATICS, 2007, 7 (03) : 331 - 368