Estimating composite functions by model selection

DOI10.1214/12-AIHP516zbMath1281.62093arXiv1102.2818MaRDI QIDQ2438264

Lucien Birgé, Yannick Baraud

Publication date: 10 March 2014

Published in: Annales de l'Institut Henri Poincaré. Probabilités et Statistiques (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1102.2818

zbMATH Keywords

model selection; artificial neural networks; Gaussian mixtures; curve estimation; adaptation; single index model

Mathematics Subject Classification ID

62G07: Density estimation

62G05: Nonparametric estimation

68T05: Learning and adaptive systems in artificial intelligence

Related Items

Estimating the conditional density by histogram type estimators and model selection, Estimating a regression function in exponential families by model selection, Estimation of the transition density of a Markov chain, Structural adaptive deconvolution under \({\mathbb{L}_p}\)-losses, Nonparametric regression using deep neural networks with ReLU activation function, \(\mathbb{L}_{p}\) adaptive estimation of an anisotropic density under independence hypothesis, Adaptive estimation over anisotropic functional classes via oracle approach, Multivariate intensity estimation via hyperbolic wavelet selection, Model selection for density estimation with \(\mathbb L_2\)-loss, Model selection for Poisson processes with covariates

Cites Work

• Unnamed Item
• Estimator selection with respect to Hellinger-type risks
• Nonparametric estimation of composite functions
• Gaussian model selection with an unknown variance
• Projection pursuit
• Risk bounds for model selection via penalization
• Approximation and estimation bounds for artificial neural networks
• Wavelet characterizations for anisotropic Besov spaces
• Model selection for Gaussian regression with random design
• Optimal global rates of convergence for nonparametric regression
• Adaptation to anisotropy and inhomogeneity via dyadic piecewise polynomial selection
• Rate-optimal estimation for a general class of nonparametric regression models with unknown link functions
• Model selection via testing: an alternative to (penalized) maximum likelihood estimators.
• Multidimensional Spline Approximation
• Universal approximation bounds for superpositions of a sigmoidal function
• Model selection for (auto-)regression with dependent data
• A Projection Pursuit Algorithm for Exploratory Data Analysis
• A non asymptotic penalized criterion for Gaussian mixture model selection
• Gaussian model selection