When there are two classes whose mean vectors and covariance matrices are known, Lanckriet et al.  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.
|Number of pages||11|
|Journal||Journal of the Operations Research Society of Japan|
|Publication status||Published - Jun 2008|
All Science Journal Classification (ASJC) codes
- Decision Sciences(all)
- Management Science and Operations Research