Estimation of Overfitting Degree of Algebraic Machine Learning in Boolean Algebra

The paper presents an estimation of overfitting probability for VKF-method of algebraic machine learning in the simplest case of Boolean algebra without counter-examples. The model uses the Vapnik—Chervonenkis proposal to minimize the empirical risk. Asymptotically the probability of overfitting err...

Full description

Saved in:
Bibliographic Details
Published inAutomatic documentation and mathematical linguistics Vol. 56; no. 3; pp. 160 - 162
Main Author Vinogradov, D.
Format Journal Article
LanguageEnglish
Published Moscow Pleiades Publishing 01.06.2022
Springer Nature B.V
Subjects
Boolean algebra
Computer Science
Fractions
Information Storage and Retrieval
Intelligent Systems
Machine learning
overfitting
empirical risk
Boolean algebra
VKF-method
Online AccessGet full text
ISSN0005-1055
1934-8371
DOI10.3103/S0005105522030098

Cover

More Information
Summary:The paper presents an estimation of overfitting probability for VKF-method of algebraic machine learning in the simplest case of Boolean algebra without counter-examples. The model uses the Vapnik—Chervonenkis proposal to minimize the empirical risk. Asymptotically the probability of overfitting errors for a fixed fraction of test examples tends to zero faster than exponentially decrease if the description length and the number of requested hypotheses go to infinity.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0005-1055
1934-8371
DOI:10.3103/S0005105522030098