Empirical risk minimization and problems of constructing linear classifiers
Source: Cybernetics and Systems Analysis, Volume 47, Number 4, July 2011 , pp. 640-648(9)
Abstract:Problems of construction of linear classifiers for classifying many sets are considered. In the case of linearly separable sets, problem statements are given that generalize already well-known formulations. For linearly inseparable sets, a natural criterion for choosing a classifier is empirical risk minimization. A mixed integer formulation of the empirical risk minimization problem and possible solutions of its continuous relaxation are considered. The proposed continuous relaxation problem is compared with problems solved with the help of other approaches to the construction of linear classifiers. Features of nonsmooth optimization methods used to solve the formulated problems are described.
Document Type: Research Article
Affiliations: 1: V. M. Glushkov Institute of Cybernetics, National Academy of Sciences of Ukraine, Kyiv, Ukraine, Email: email@example.com 2: A. A. Dorodnitsyn Computing Center, Russian Academy of Sciences, Moscow, Russia, Email: firstname.lastname@example.org 3: A. A. Dorodnitsyn Computing Center, Russian Academy of Sciences, Moscow, Russia, Email: email@example.com
Publication date: July 1, 2011