Email this article Entropy Dimension Reduction Method for Randomized Machine Learning Problems
- Authors: Popkov Y.S.1,2,3, Dubnov Y.A.1,3,4, Popkov A.Y.1,5
-
Affiliations:
- Institute for Systems Analysis, Russian Academy of Sciences
- Braude College of Haifa University
- National Research University “Higher School of Economics,”
- Moscow Institute of Physics and Technology
- Peoples’ Friendship University
- Issue: Vol 79, No 11 (2018)
- Pages: 2038-2051
- Section: Control in Technical Systems
- URL: https://journals.rcsi.science/0005-1179/article/view/151074
- DOI: https://doi.org/10.1134/S0005117918110085
- ID: 151074
Cite item
Open Access
Access granted
Subscription Access
Abstract
The direct and inverse projections (DIP) method was proposed to reduce the feature space to the given dimensions oriented to the problems of randomized machine learning and based on the procedure of “direct” and “inverse” design. The “projector” matrices are determined by maximizing the relative entropy. It is suggested to estimate the information losses by the absolute error calculated with the use of the Kullback–Leibler function (SRC method). An example illustrating these methods was given.
About the authors
Yu. S. Popkov
Institute for Systems Analysis, Russian Academy of Sciences; Braude College of Haifa University; National Research University “Higher School of Economics,”
Author for correspondence.
Email: popkov@isa.ru
Russian Federation, Moscow; Carmiel; Moscow
Yu. A. Dubnov
Institute for Systems Analysis, Russian Academy of Sciences; National Research University “Higher School of Economics,”; Moscow Institute of Physics and Technology
Email: popkov@isa.ru
Russian Federation, Moscow; Moscow; Moscow
A. Yu. Popkov
Institute for Systems Analysis, Russian Academy of Sciences; Peoples’ Friendship University
Email: popkov@isa.ru
Russian Federation, Moscow; Moscow
Supplementary files
