Incremental DC optimization algorithm for large-scale clusterwise linear regression

被引:4
|
作者
Bagirov, Adil M. [1 ]
Taheri, Sona [2 ]
Cimen, Emre [3 ]
机构
[1] Federat Univ Australia, Sch Engn Informat Technol & Phys Sci, Univ Dr, Mt Helen, Vic 3350, Australia
[2] RMIT Univ, Sch Math Sci, Melbourne, Vic, Australia
[3] Eskisehir Tech Univ, Dept Ind Engn, Iki Eylul Campus, TR-26555 Eskisehir, Turkey
基金
澳大利亚研究理事会;
关键词
Regression analysis; Clusterwise linear regression; Nonsmooth optimization; Nonconvex optimization; DC optimization;
D O I
10.1016/j.cam.2020.113323
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
The objective function in the nonsmooth optimization model of the clusterwise linear regression (CLR) problem with the squared regression error is represented as a difference of two convex functions. Then using the difference of convex algorithm (DCA) approach the CLR problem is replaced by the sequence of smooth unconstrained optimization subproblems. A new algorithm based on the DCA and the incremental approach is designed to solve the CLR problem. We apply the Quasi-Newton method to solve the subproblems. The proposed algorithm is evaluated using several synthetic and real world data sets for regression and compared with other algorithms for CLR. Results demonstrate that the DCA based algorithm is efficient for solving CLR problems with the large number of data points and in particular, outperforms other algorithms when the number of input variables is small. (C) 2020 Elsevier B.V. All rights reserved.
引用
收藏
页数:17
相关论文
共 50 条
  • [1] Nonsmooth Optimization Algorithm for Solving Clusterwise Linear Regression Problems
    Adil M. Bagirov
    Julien Ugon
    Hijran G. Mirzayeva
    [J]. Journal of Optimization Theory and Applications, 2015, 164 : 755 - 780
  • [2] Nonsmooth Optimization Algorithm for Solving Clusterwise Linear Regression Problems
    Bagirov, Adil M.
    Ugon, Julien
    Mirzayeva, Hijran G.
    [J]. JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2015, 164 (03) : 755 - 780
  • [3] DC Programming Algorithm for Clusterwise Linear L1 Regression
    Bagirov, Adil M.
    Taheri, Sona
    [J]. Journal of the Operations Research Society of China, 2017, 5 (02) : 233 - 256
  • [4] Incremental learning algorithm for large-scale semi-supervised ordinal regression
    Chen, Haiyan
    Jia, Yizhen
    Ge, Jiaming
    Gu, Bin
    [J]. NEURAL NETWORKS, 2022, 149 : 124 - 136
  • [5] A FAST ALGORITHM FOR CLUSTERWISE LINEAR-REGRESSION
    SPATH, H
    [J]. COMPUTING, 1982, 29 (02) : 175 - 181
  • [6] Clusterwise linear regression modeling with soft scale constraints
    Di Mari, Roberto
    Rocci, Roberto
    Gattone, Stefano Antonio
    [J]. INTERNATIONAL JOURNAL OF APPROXIMATE REASONING, 2017, 91 : 160 - 178
  • [7] An algorithm for clusterwise linear regression based on smoothing techniques
    Bagirov, Adil M.
    Ugon, Julien
    Mirzayeva, Hijran G.
    [J]. OPTIMIZATION LETTERS, 2015, 9 (02) : 375 - 390
  • [8] A FAST ALGORITHM FOR CLUSTERWISE LINEAR ABSOLUTE DEVIATIONS REGRESSION
    MEIER, J
    [J]. OR SPEKTRUM, 1987, 9 (03) : 187 - 189
  • [9] An algorithm for clusterwise linear regression based on smoothing techniques
    Adil M. Bagirov
    Julien Ugon
    Hijran G. Mirzayeva
    [J]. Optimization Letters, 2015, 9 : 375 - 390
  • [10] Large-scale Linear Support Vector Regression
    Ho, Chia-Hua
    Lin, Chih-Jen
    [J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2012, 13 : 3323 - 3348