QUADRATIC AND CONVEX MINIMAX CLASSIFICATION PROBLEMS

• Kitahara Tomonari

  Tokyo Institute of Technology

• Mizuno Shinji

  Tokyo Institute of Technology

• Nakata Kazuhide

  Tokyo Institute of Technology

抄録

When there are two classes whose mean vectors and covariance matrices are known, Lanckriet et al. [7] consider the Linear Minimax Classification (LMC) problem and they propose a method for solving it. In this paper we first discuss the Quadratic Minimax Classification (QMC) problem, which is a generalization of LMC. We show that QMC is transformed to a parametric Semidefinite Programming (SDP) problem. We further define the Convex Minimax Classification (CMC) problem. Though the two problems are generalizations of LMC, we prove that solutions of these problems can be obtained by solving LMC.

収録刊行物

• 日本オペレーションズ・リサーチ学会論文誌

  日本オペレーションズ・リサーチ学会論文誌 51 (2), 191-201, 2008

  公益社団法人 日本オペレーションズ・リサーチ学会

キーワード

• Optimization
• minimax classification
• semidefinite programming
• convexity

詳細情報

• CRID: 1390282679085901056
• NII論文ID: 110006792052
• NII書誌ID: AA00703935
• DOI: 10.15807/jorsj.51.191
• ISSN: 21888299
• ISSN: 04534514
• NDL書誌ID: 9544559
• Web Site: https://ndlsearch.ndl.go.jp/books/R000000004-I9544559
• Web Site: https://www.jstage.jst.go.jp/article/jorsj/51/2/51_KJ00004960694/_pdf
• 本文言語コード: en
• データソース種別:
  • JaLC
  • NDL
  • Crossref
  • CiNii Articles
  • KAKEN
• 抄録ライセンスフラグ: 使用不可