Efficient agnostic learning of neural networks with bounded fan-in

DOI10.1109/18.556601zbMath0874.68253OpenAlexW2028461624MaRDI QIDQ4336393

Bartlett, Peter L., Wee Sun Lee, Robert C. Williamson

Publication date: 12 June 1997

Published in: IEEE Transactions on Information Theory (Search for Journal in Brave)

Full work available at URL: https://semanticscholar.org/paper/2f41eedb489db10ce8e9a469931f0d1741c669e4

zbMATH Keywords

learning modeltwo-layer neural networksrealistic extension

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05)

Related Items (31)

Deep learning: a statistical viewpoint ⋮ Local Rademacher complexities and oracle inequalities in risk minimization. (2004 IMS Medallion Lecture). (With discussions and rejoinder) ⋮ Rescaled pure greedy algorithm for Hilbert and Banach spaces ⋮ Benign overfitting in linear regression ⋮ Nonlinear function approximation: computing smooth solutions with an adaptive greedy algorithm ⋮ Unnamed Item ⋮ Best subset selection, persistence in high-dimensional statistical learning and optimization under \(l_1\) constraint ⋮ Multi-kernel regularized classifiers ⋮ Nonlinear orthogonal series estimates for random design regression ⋮ Nonexact oracle inequalities, \(r\)-learnability, and fast rates ⋮ Deep nonparametric regression on approximate manifolds: nonasymptotic error bounds with polynomial prefactors ⋮ Greedy training algorithms for neural networks and applications to PDEs ⋮ Generalization Analysis of Fredholm Kernel Regularized Classifiers ⋮ Agnostic Learning from Tolerant Natural Proofs ⋮ CONVERGENCE OF A LEAST‐SQUARES MONTE CARLO ALGORITHM FOR AMERICAN OPTION PRICING WITH DEPENDENT SAMPLE DATA ⋮ Gradient Descent with Identity Initialization Efficiently Learns Positive-Definite Linear Transformations by Deep Residual Networks ⋮ Learning by mirror averaging ⋮ Approximation and learning by greedy algorithms ⋮ Persistene in high-dimensional linear predictor-selection and the virtue of overparametrization ⋮ A note on margin-based loss functions in classification ⋮ Monte Carlo algorithms for optimal stopping and statistical learning ⋮ Large-Margin Classification in Infinite Neural Networks ⋮ Boosting the margin: a new explanation for the effectiveness of voting methods ⋮ Scale-sensitive dimensions and skeleton estimates for classification ⋮ General Error Estimates for the Longstaff–Schwartz Least-Squares Monte Carlo Algorithm ⋮ The complexity of model classes, and smoothing noisy data ⋮ Inequalities for uniform deviations of averages from expectations with applications to nonparametric regression ⋮ Functional aggregation for nonparametric regression. ⋮ Local greedy approximation for nonlinear regression and neural network training. ⋮ Boosting with early stopping: convergence and consistency ⋮ Hardness results for neural network approximation problems

This page was built for publication: Efficient agnostic learning of neural networks with bounded fan-in