Complexity regularization via localized random penalties

DOI10.1214/009053604000000463zbMath1045.62060arXivmath/0410091OpenAlexW3098965260MaRDI QIDQ1879970

Gábor Lugosi, Marten H. Wegkamp

Publication date: 15 September 2004

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/math/0410091

zbMATH Keywords

classificationconcentration inequalitiesoracle inequalitiescomplexity regularizationRademacher averagesrandom penaltiesshatter coefficients

Mathematics Subject Classification ID

Classification and discrimination; cluster analysis (statistical aspects) (62H30) Inequalities; stochastic orderings (60E15) Nonparametric inference (62G99)

Related Items

Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder), Aggregation via empirical risk minimization, Classifiers of support vector machine type with \(\ell_1\) complexity regularization, Model selection in reinforcement learning, Statistical performance of support vector machines, Bootstrap model selection for possibly dependent and heterogeneous data, Model selection by resampling penalization, Margin-adaptive model selection in statistical learning, Adaptive kernel methods using the balancing principle, Bayesian fractional posteriors, Simultaneous adaptation to the margin and to complexity in classification, Learning by mirror averaging, Empirical minimization, Sparse estimation by exponential weighting, Generalized mirror averaging and \(D\)-convex aggregation, Theory of Classification: a Survey of Some Recent Advances, On the Optimality of Sample-Based Estimates of the Expectation of the Empirical Minimizer, Rademacher complexity in Neyman-Pearson classification, Square root penalty: Adaption to the margin in classification and in edge estimation, Local Rademacher complexities

Cites Work

• Some limit theorems for empirical processes (with discussion)
• A result of Vapnik with applications
• Concentration inequalities using the entropy method
• Adaptive model selection using empirical complexities
• A Bennett concentration inequality and its application to suprema of empirical processes
• Empirical margin distributions and bounding the generalization error of combined classifiers
• About the constants in Talagrand's concentration inequalities for empirical processes.
• New approaches to statistical learning theory
• Optimal aggregation of classifiers in statistical learning.
• Weak convergence and empirical processes. With applications to statistics
• A new look at independence
• On Talagrand's deviation inequalities for product measures
• Theorie der Zeichenerkennung
• A sharp concentration inequality with applications
• Rademacher penalties and structural risk minimization
• Structural risk minimization over data-dependent hierarchies
• Some applications of concentration inequalities to statistics
• Concentration inequalities for set-indexed empirical processes
• Model selection and error estimation
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item