DCA-based algorithms for DC fitting

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS

卷:389

Article

Vinh Thanh Ho^1,2  Hoai An Le Thi³  Tao Pham Dinh⁴ 

[1] Ton Duc Thang Univ, Dept Management Sci & Technol Dev, Ho Chi Minh City, Vietnam

[2] Ton Duc Thang Univ, Fac Math & Stat, Ho Chi Minh City, Vietnam

[3] Univ Lorraine, LGIPM, F-57000 Metz, France

[4] Univ Normandie, INSA Rouen, Lab Math, F-76801 St Etienne Du Rouvray, France

关键词: DC programming;    DCA;    DCA with successive DC decomposition;    DC fitting;    Piecewise-linear fitting;   

DOI  :  10.1016/j.cam.2020.113353

来源: Elsevier

【 摘 要 】

We investigate a nonconvex, nonsmooth optimization approach based on DC (Difference of Convex functions) programming and DCA (DC Algorithm) for the so-called DC fitting problem, which aims to fit a given set of data points by a DC function. The problem is tackled as minimizing the squared Euclidean norm fitting error. It is formulated as a DC program for which a standard DCA scheme is developed. Furthermore, a modified DCA scheme with successive DC decomposition is proposed. These standard/modified versions of DCA are applied for solving the continuous piecewise-linear fitting problem. Numerical experiments on many synthetic and real datasets with small-to-large sizes show the efficiency of our DCA-based approach in comparison with the existing approaches for constructing continuous piecewise-linear models. (C) 2020 Elsevier B.V. All rights reserved.

【 授权许可】

Free