A Two-Stage Method for Constructing Linear Regressions Using Optimal Convex Combinations
- Авторы: Senko O.1, Dokukin A.1, Kiselyova N.2, Khomutov N.1
-
Учреждения:
- Dorodnicyn Computing Center, Federal Research Center “Computer Science and Control”
- Baikov Institute of Metallurgy and Materials Science
- Выпуск: Том 97, № 2 (2018)
- Страницы: 113-114
- Раздел: Mathematics
- URL: https://journals.rcsi.science/1064-5624/article/view/225470
- DOI: https://doi.org/10.1134/S1064562418020035
- ID: 225470
Цитировать
Аннотация
Multilevel learning systems have become more popular in pattern recognition and regression analysis. In this paper, a two-level method for constructing a multidimensional regression model is considered, in which a family of optimal convex combinations of simple one-dimensional least-square regressions is generated at the first level. The second level of the proposed learning system is given by an elastic net. Experimental verification presented demonstrate the efficiency of the proposed regression estimation method as applied to problems with a small amount of data.
Об авторах
O. Senko
Dorodnicyn Computing Center, Federal Research Center “Computer Science and Control”
Email: dalex@ccas.ru
Россия, Moscow, 119333
A. Dokukin
Dorodnicyn Computing Center, Federal Research Center “Computer Science and Control”
Автор, ответственный за переписку.
Email: dalex@ccas.ru
Россия, Moscow, 119333
N. Kiselyova
Baikov Institute of Metallurgy and Materials Science
Email: dalex@ccas.ru
Россия, Moscow, 119991
N. Khomutov
Dorodnicyn Computing Center, Federal Research Center “Computer Science and Control”
Email: dalex@ccas.ru
Россия, Moscow, 119333