Bound estimation-based safe acceleration for maximum margin of twin spheres machine with pinball loss

被引:15
作者
Yuan, Min [1 ]
Xu, Yitian [2 ]
机构
[1] China Agr Univ, Coll Informat & Elect Engn, Beijing 100083, Peoples R China
[2] China Agr Univ, Coll Sci, Beijing 100083, Peoples R China
基金
中国国家自然科学基金; 北京市自然科学基金;
关键词
Maximum margin; Pinball loss; Imbalanced data; Bound estimation; Upper and lower bounds; SUPPORT VECTOR MACHINE; CLASSIFIER;
D O I
10.1016/j.patcog.2021.107860
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Maximum margin of twin spheres support vector machine (MMTSSVM) is an efficient method for imbalanced data classification. As an extension to enhance noise insensitivity of MMTSSVM, MMTSSVM with pinball loss (Pin-MMTSM) has a good generalization performance. However, it is not efficient enough for large-scale data. Inspired by the sparse solution of SVMs, in this paper, we propose a safe accelerative approach to reduce the computational cost. Unlike the existing safe screening rules, where only one variable changes with the parameters. We utilize bound estimation-based to derive the upper and lower bounds of center and radius. With our approach, the inactive samples are discarded before solving the problem, thus it can reduce the computational cost. One important advantage of our approach is safety, i.e., we can obtain the same solution as solving original problem both in linear and non-linear cases. Moreover, it is obvious that our acceleration approach is independent of the solver. To further accelerate the computational speed, a decomposition method is employed. Experiments on three artificial datasets and twelve benchmark datasets clearly demonstrate the effectiveness of our approach. At last, we extend bound estimation-based method to nu-SVM, theoretical analysis and experimental results both verify its feasibility and effectiveness. (C) 2021 Elsevier Ltd. All rights reserved.
引用
收藏
页数:11
相关论文
共 40 条
[1]  
[Anonymous], 2013, P 30 INT C MACH LEAR
[2]   Soft clustering using weighted one-class support vector machines [J].
Bicego, Manuele ;
Figueiredo, Mario A. T. .
PATTERN RECOGNITION, 2009, 42 (01) :27-32
[3]   Dynamic Screening: Accelerating First-Order Algorithms for the Lasso and Group-Lasso [J].
Bonnefoy, Antoine ;
Emiya, Valentin ;
Ralaivola, Liva ;
Gribonval, Remi .
IEEE TRANSACTIONS ON SIGNAL PROCESSING, 2015, 63 (19) :5121-5132
[4]  
Chang C., 2013, KNOWL-BASED SYST, V43, P74
[5]   Training ν-support vector classifiers:: Theory and algorithms [J].
Chang, CC ;
Lin, CJ .
NEURAL COMPUTATION, 2001, 13 (09) :2119-2147
[6]  
CORTES C, 1995, MACH LEARN, V20, P273, DOI 10.1023/A:1022627411411
[7]   DCSVM: fast multi-class classification using support vector machines [J].
Don, Duleep Rathgamage ;
Iacob, Ionut E. .
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2020, 11 (02) :433-447
[8]  
El Ghaoui L, 2012, PAC J OPTIM, V8, P667
[9]   1-Norm least squares twin support vector machines [J].
Gao, Shangbing ;
Ye, Qiaolin ;
Ye, Ning .
NEUROCOMPUTING, 2011, 74 (17) :3590-3597
[10]  
Guler O, 2010, GRAD TEXTS MATH, V258, P1, DOI 10.1007/978-0-387-68407-9