Generalization properties of finite size polynomial Support Vector Machines

Дата и время публикации : 2000-02-04T16:40:28Z

Авторы публикации и институты :
Sebastian Risau-Gusman
Mirta B. Gordon

Ссылка на журнал-издание: Ссылка на журнал-издание не найдена
Коментарии к cтатье: 12 pages, 7 figures
Первичная категория: cond-mat.dis-nn

Все категории : cond-mat.dis-nn, cond-mat.stat-mech

Краткий обзор статьи: The learning properties of finite size polynomial Support Vector Machines are analyzed in the case of realizable classification tasks. The normalization of the high order features acts as a squeezing factor, introducing a strong anisotropy in the patterns distribution in feature space. As a function of the training set size, the corresponding generalization error presents a crossover, more or less abrupt depending on the distribution’s anisotropy and on the task to be learned, between a fast-decreasing and a slowly decreasing regime. This behaviour corresponds to the stepwise decrease found by Dietrich et al.[Phys. Rev. Lett. 82 (1999) 2975-2978] in the thermodynamic limit. The theoretical results are in excellent agreement with the numerical simulations.

Category: Physics