Computational complexity of combinatorial optimization problems induced by collective procedures in machine learning / Khachai M. Yu. // PROCEEDINGS OF THE STEKLOV INSTITUTE OF MATHEMATICS. - 2011. - V. 272, l. 1. - P. 46-54.

ISSN/EISSN:

0081-5438 / нет данных

Type:

Article

Abstract:

The computational complexity of a new class of combinatorial optimization problems that are induced by optimal machine learning procedures in the class of collective piecewise linear classifiers of committee type is studied.

Author keywords:

empirical risk minimization; committee classifier; computational complexity

DOI:

10.1134/S0081543811020040

Web of Science ID:

ISI:000289527400004

Соавторы в МНС:

—

Другие поля

Поле	Значение
Month	APR
Publisher	MAIK NAUKA/INTERPERIODICA/SPRINGER
Address	233 SPRING ST, NEW YORK, NY 10013-1578 USA
Language	English
Research-Areas	Mathematics
Web-of-Science-Categories	Mathematics, Applied; Mathematics
Author-Email	mkhachay@imm.uran.ru
ResearcherID-Numbers	Khachay, Michael/H-3251-2013
ORCID-Numbers	Khachay, Michael/0000-0003-3555-0080
Funding-Acknowledgement	Russian Foundation for Basic Research {[}07-07-00168, 09-01-00139]; Presidium of the Ural Branch of the Russian Academy of Sciences {[}09-P-1-1001, 09-S-1-1010]
Funding-Text	This work was supported by the Russian Foundation for Basic Research (project nos. 07-07-00168 and 09-01-00139) and by the Presidium of the Ural Branch of the Russian Academy of Sciences (project nos. 09-P-1-1001 and 09-S-1-1010).
Number-of-Cited-References	14
Journal-ISO	Proc. Steklov Inst. Math.
Doc-Delivery-Number	750EJ