Any Discrimination Rule Can Have an Arbitrarily Bad Probability of Error for Finite Sample Size

DOI10.1109/TPAMI.1982.4767222zbMath0484.62072MaRDI QIDQ3943843

Luc P. Devroye

Publication date: 1982

Published in: IEEE Transactions on Pattern Analysis and Machine Intelligence (Search for Journal in Brave)

zbMATH Keywords

Bayes risk; distribution-free inequalities

Mathematics Subject Classification ID

62H30: Classification and discrimination; cluster analysis (statistical aspects)

62C10: Bayesian problems; characterization of Bayes procedures

Related Items

Consistency of support vector machines using additive kernels for additive models, Asymptotic normality of support vector machine variants and other regularized kernel methods, Optimal global rates of convergence for nonparametric regression with unbounded data, Lower bounds for the rate of convergence in nonparametric pattern recognition, Bandwidth choice for nonparametric classification, Asymptotic expansions of the \(k\) nearest neighbor risk, Nonlinear black-box models in system identification: Mathematical foundations, Lower bounds on the rate of convergence of nonparametric regression estimates, Fast learning rates in statistical inference through aggregation, Rates of convergence for partitioning and nearest neighbor regression estimates with unbounded data, Convergence properties of functional estimates for discrete distributions