Sharpness estimation of combinatorial generalization ability bounds for threshold decision rules

DOI10.1134/S0005117921050106MaRDI QIDQ2034842zbMATH OpenOpenAlexFDO

Authors Sh. Kh. Ishkina, K. V. Vorontsov

Publication date 23 June 2021

Published in Automation and Remote Control (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1134/s0005117921050106

zbMATH Keywords

Rademacher complexitygeneralization abilitycombinatorial theorycomplete cross-validationprobability of overfittingthreshold classifier

Mathematics Subject Classification ID

Decision theory (91B06)

Recommendations

• Combinatorial bounds of overfitting for threshold classifiers
• scientific article; zbMATH DE number 2206866
• Combinatorial bounds for learning performance
• Generalization error bounds for threshold decision lists
• Generalization error of combined classifiers.

Cites work

• scientific article; zbMATH DE number 3664426 (Why is no real title available?)
• scientific article; zbMATH DE number 3446442 (Why is no real title available?)
• A decision-theoretic generalization of on-line learning and an application to boosting
• Combinatorial bounds of overfitting for threshold classifiers
• On the Uniform Convergence of Relative Frequencies of Events to Their Probabilities
• Oracle inequalities in empirical risk minimization and sparse recovery problems. École d'Été de Probabilités de Saint-Flour XXXVIII-2008.
• Predicting \(\{ 0,1\}\)-functions on randomly drawn points
• Probability Inequalities for Sums of Bounded Random Variables
• Theory of Classification: a Survey of Some Recent Advances
• Tight bounds for the probability of overfitting
• Understanding machine learning. From theory to algorithms

Cited in

(1)

• Combinatorial bounds of overfitting for threshold classifiers

Describes a project that uses

Uses Software

• GitHub
• AdaBoost.MH

This page was built for publication: Sharpness estimation of combinatorial generalization ability bounds for threshold decision rules

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2034842)