Large margin classifiers based on affine hulls

被引:34
|
作者
Cevikalp, Hakan [1 ]
Triggs, Bill [2 ]
Yavuz, Hasan Serhan [1 ]
Kucuk, Yalcin [3 ]
Kucuk, Mahide [3 ]
Barkana, Atalay [4 ]
机构
[1] Eskisehir Osmangazi Univ, Dept Elect & Elect Engn, TR-26480 Meselik, Eskisehir, Turkey
[2] Lab Jean Kuntzmann, Al Apprentissage & Interfaces Team, Grenoble, France
[3] Anadolu Univ, Dept Math, Eskisehir, Turkey
[4] Anadolu Univ, Dept Elect & Elect Engn, Eskisehir, Turkey
关键词
Affine hull; Classification; Convex hull; Kernel methods; Large margin classifier; Quadratic programming; Support vector machines; SUPPORT VECTOR MACHINES; ALGORITHMS; MANIFOLDS;
D O I
10.1016/j.neucom.2010.06.018
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper introduces a geometrically inspired large margin classifier that can be a better alternative to the support vector machines (SVMs) for the classification problems with limited number of training samples. In contrast to the SVM classifier, we approximate classes with affine hulls of their class samples rather than convex hulls. For any pair of classes approximated with affine hulls, we introduce two solutions to find the best separating hyperplane between them. In the first proposed formulation, we compute the closest points on the affine hulls of classes and connect these two points with a line segment. The optimal separating hyperplane between the two classes is chosen to be the hyperplane that is orthogonal to the line segment and bisects the line. The second formulation is derived by modifying the v SVM formulation. Both formulations are extended to the nonlinear case by using the kernel trick. Based on our findings, we also develop a geometric interpretation of the least squares SVM classifier and show that it is a special case of the proposed method. Multi-class classification problems are dealt with constructing and combining several binary classifiers as in SVM. The experiments on several databases show that the proposed methods work as good as the SVM classifier if not any better. (C) 2010 Elsevier B.V. All rights reserved.
引用
收藏
页码:3160 / 3168
页数:9
相关论文
共 50 条
  • [1] Geometric Algorithms to Large Margin Classifier Based on Affine Hulls
    Peng, Xinjun
    Wang, Yifei
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2012, 23 (02) : 236 - 246
  • [2] Incremental margin algorithm for large margin classifiers
    Leite, Saul C.
    Neto, Raul Fonseca
    NEUROCOMPUTING, 2008, 71 (7-9) : 1550 - 1560
  • [3] Adaptively Weighted Large Margin Classifiers
    Wu, Yichao
    Liu, Yufeng
    JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2013, 22 (02) : 416 - 432
  • [4] Multiplicative updates for large margin classifiers
    Sha, F
    Saul, LK
    Lee, DD
    LEARNING THEORY AND KERNEL MACHINES, 2003, 2777 : 188 - 202
  • [5] Convex Bidirectional Large Margin Classifiers
    Qi, Zhengling
    Liu, Yufeng
    TECHNOMETRICS, 2019, 61 (02) : 176 - 186
  • [6] Large margin nearest neighbor classifiers
    Domeniconi, C
    Gunopulos, D
    Peng, J
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 2005, 16 (04): : 899 - 909
  • [7] Apportioned margin approach for cost sensitive large margin classifiers
    Gottlieb, Lee-Ad
    Kaufman, Eran
    Kontorovich, Aryeh
    ANNALS OF MATHEMATICS AND ARTIFICIAL INTELLIGENCE, 2021, 89 (12) : 1215 - 1235
  • [8] Apportioned margin approach for cost sensitive large margin classifiers
    Lee-Ad Gottlieb
    Eran Kaufman
    Aryeh Kontorovich
    Annals of Mathematics and Artificial Intelligence, 2021, 89 : 1215 - 1235
  • [9] Adaptively weighted large-margin angle-based classifiers
    Fu, Sheng
    Zhang, Sanguo
    Liu, Yufeng
    JOURNAL OF MULTIVARIATE ANALYSIS, 2018, 166 : 282 - 299
  • [10] An approach to large margin design of prototype-based pattern classifiers
    He, Tingting
    Hu, Yu
    Huo, Qiang
    2007 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOL II, PTS 1-3, 2007, : 625 - +