Entropy Dimension Reduction Method for Randomized Machine Learning Problems

被引：3

作者：

Popkov, Yu. S. ^{[1
,2
]}

Dubnov, Yu. A. ^{[1
,3
,4
]}

Popkov, A. Yu. ^{[1
,3
,5
]}

机构：

[1] Russian Acad Sci, Fed Res Ctr Informat & Control, Inst Syst Anal, Moscow, Russia

[2] Univ Haifa, Braude Coll, Carmiel, Israel

[3] Natl Res Univ, Higher Sch Econ, Moscow, Russia

[4] Moscow Inst Phys & Technol, Moscow, Russia

[5] Peoples Friendship Univ, Moscow, Russia

来源：

AUTOMATION AND REMOTE CONTROL | 2018年 / 79卷 / 11期

基金：

俄罗斯基础研究基金会;

关键词：

entropy; relative entropy; projection operators; matrix derivatives; gradient method; direct and inverse projections; INFORMATION; ALGORITHMS;

D O I：

10.1134/S0005117918110085

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

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.

引用

页码：2038 / 2051

页数：14

共 50 条

[1] Entropy Dimension Reduction Method for Randomized Machine Learning Problems
Yu. S. Popkov
Yu. A. Dubnov
A. Yu. Popkov
Automation and Remote Control, 2018, 79 : 2038 - 2051
[2] Dimension Reduction With Extreme Learning Machine
Kasun, Liyanaarachchi Lekamalage Chamara
Yang, Yan
Huang, Guang-Bin
Zhang, Zhengyou
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2016, 25 (08) : 3906 - 3918
[3] A method for dimension reduction in quadratic classification problems
Velilla, Santiago
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS, 2008, 17 (03) : 572 - 589
[4] NORMALIZATION AND DIMENSION REDUCTION FOR MACHINE LEARNING IN ADVANCED MANUFACTURING
Huang, Jida
Kwok, Tsz-Ho
PROCEEDINGS OF ASME 2022 INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, IDETC-CIE2022, VOL 2, 2022,
[5] A Weighting Method for Feature Dimension by Semisupervised Learning With Entropy
Jin, Dequan
Yang, Murong
Qin, Ziyan
Peng, Jigen
Ying, Shihui
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2023, 34 (03) : 1218 - 1227
[6] PMV Dimension Reduction Utilizing Feature Selection Method: Comparison Study on Machine Learning Models
Park, Kyung-Yong
Woo, Deok-Oh
ENERGIES, 2023, 16 (05)
[7] A combined reliability analysis approach with dimension reduction method and maximum entropy method
Gang Li
Kai Zhang
Structural and Multidisciplinary Optimization, 2011, 43 : 121 - 134
[8] A combined reliability analysis approach with dimension reduction method and maximum entropy method
Li, Gang
Zhang, Kai
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION, 2011, 43 (01) : 121 - 134
[9] Comparing Swarm Intelligence Algorithms for Dimension Reduction in Machine Learning
Kicska, Gabriella
Kiss, Attila
BIG DATA AND COGNITIVE COMPUTING, 2021, 5 (03)
[10] A novel dimension reduction method with information entropy to evaluate network resilience
Wu, Chengxing
Duan, Dongli
Xiao, Renbin
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS, 2023, 620

← 1 2 3 4 5 →