Bounds on the sample complexity of Bayesian learning using information theory and the VC dimension

zbMath0798.68145MaRDI QIDQ1314506

David Haussler, Michael Kearns, Robert E. Schapire

Publication date: 3 March 1994

Published in: Machine Learning (Search for Journal in Brave)

zbMATH Keywords

Bayesian learningVC dimensionstatistical physicsinformation theorylearning curvesaverage-case learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (11)

Sphere packing numbers for subsets of the Boolean \(n\)-cube with bounded Vapnik-Chervonenkis dimension ⋮ Learning from a population of hypotheses ⋮ Characterizing rational versus exponential learning curves ⋮ Mutual information, metric entropy and cumulative relative entropy risk ⋮ MetaBayes: Bayesian Meta-Interpretative Learning Using Higher-Order Stochastic Refinement ⋮ Rigorous learning curve bounds from statistical mechanics ⋮ Learning a priori constrained weighted majority votes ⋮ QG/GA: a stochastic search for Progol ⋮ Sample size lower bounds in PAC learning by Algorithmic Complexity Theory ⋮ Bayesian predictiveness, exchangeability and sufficientness in bacterial taxonomy ⋮ Query by committee, linear separation and random walks.

This page was built for publication: Bounds on the sample complexity of Bayesian learning using information theory and the VC dimension