中国邮电高校学报(英文) ›› 2020, Vol. 27 ›› Issue (3): 31-41.doi: 10.19682/j.cnki.1005-8885.2020.0014

• Artificial Intelligence • 上一篇    下一篇

Application of smoothing technique on twin support vector hypersphere

吴青1,高小凤2,范九伦3,张恒昌1   

  1. 1. 西安邮电大学
    2.
    3. 西安邮电学院
  • 收稿日期:2019-11-08 修回日期:2020-04-12 出版日期:2020-06-24 发布日期:2020-08-30
  • 通讯作者: 吴青 E-mail:xiyouwuq@126.com
  • 基金资助:
    国家自然科学基金;陕西重点科研项目;陕西国际科技合作计划;陕西教育部基金会

Application of smoothing technique on twin support vector hypersphere

  • Received:2019-11-08 Revised:2020-04-12 Online:2020-06-24 Published:2020-08-30
  • Contact: Qing Wu E-mail:xiyouwuq@126.com
  • Supported by:
    National Natural Science Foundation of China;the Key Research Project of Shanxi Province;the International S&T Cooperation Program of Shanxi Province;the Foundation of Education Department of Shanxi Province

摘要: In order to improve the learning speed and reduce computational complexity of twin support vector hypersphere (TSVH), this paper presents a smoothed twin support vector hypersphere (STSVH) based on the smoothing technique. STSVH can generate two hyperspheres with each one covering as many samples as possible from the same class respectively. Additionally, STSVH only solves a pair of unconstraint differentiable quadratic programming problems (QPPs) rather than a pair of constraint dual QPPs which makes STSVH faster than the TSVH. By considering the differentiable characteristics of STSVH, a fast Newton-Armijo algorithm is used for solving STSVH. Numerical experiment results on normally distributed clustered datasets ( NDC) as well as University of California Irvine (UCI) data sets indicate that the significant advantages of the proposed STSVH in terms of efficiency and generalization performance.

关键词: twin support vector hypersphere, Newton-Armijo algorithm, smoothing approximation function, unconstraint differentiable optimization

Abstract: In order to improve the learning speed and reduce computational complexity of twin support vector hypersphere (TSVH), this paper presents a smoothed twin support vector hypersphere (STSVH) based on the smoothing technique. STSVH can generate two hyperspheres with each one covering as many samples as possible from the same class respectively. Additionally, STSVH only solves a pair of unconstraint differentiable quadratic programming problems (QPPs) rather than a pair of constraint dual QPPs which makes STSVH faster than the TSVH. By considering the differentiable characteristics of STSVH, a fast Newton-Armijo algorithm is used for solving STSVH. Numerical experiment results on normally distributed clustered datasets ( NDC) as well as University of California Irvine (UCI) data sets indicate that the significant advantages of the proposed STSVH in terms of efficiency and generalization performance.

Key words: twin support vector hypersphere, Newton-Armijo algorithm, smoothing approximation function, unconstraint differentiable optimization