Rates of contraction with respect to L₂-distance for Bayesian nonparametric regression
From MaRDI portal
Publication:2326065
Abstract: We develop a unifying framework for Bayesian nonparametric regression to study the rates of contraction with respect to the integrated -distance without assuming the regression function space to be uniformly bounded. The framework is very flexible and can be applied to a wide class of nonparametric prior models. Three non-trivial applications of the proposed framework are provided: The finite random series regression of an -H"older function, with adaptive rates of contraction up to a logarithmic factor; The un-modified block prior regression of an -Sobolev function, with adaptive-and-exact rates of contraction; The Gaussian spline regression of an -H"older function, with the near-optimal posterior contraction. These applications serve as generalization or complement of their respective results in the literature. Extensions to the fixed-design regression problem and sparse additive models in high dimensions are discussed as well.
This paper is about the problem of rates of convergence in nonparametric regression, in a version that applies to posterior distributions. The point of departure is the work [Ann. Stat. 28, No. 2, 500--531 (2000; Zbl 1105.62315)] by \textit{S. Ghosal} et al., where a framework of prior concentration, sieves and test functions is set up in order to approach this problem. According to the authors of the paper under review, the ``major contribution of this work \(\ldots\) is that we impose certain structural assumption(s) on the sieves to construct the desired test functions. By doing so, we are able to make the framework more concrete and instructive for a variety of nonparametric regression priors \(\dots\) (p. 3489). It is also emphasized that the space of regression functions is allowed to be unbounded.
Recommendations
- On Bayesian supremum norm contraction rates
- scientific article; zbMATH DE number 721749
- Nonparametric Bayesian posterior contraction rates for discretely observed scalar diffusions
- Convergence rates for empirical bayes estimators of parameters in linear regressin models
- Rates of contraction for posterior distributions in \(L^{r}\)-metrics, \(1 \leq r \leq \infty\)
- Lower bounds on the rate of convergence of nonparametric regression estimates
- scientific article; zbMATH DE number 1116662
- scientific article; zbMATH DE number 1064648
Cites work
- A practical guide to splines
- A theoretical framework for calibration in computer models: parametrization, estimation and convergence properties
- Adaptive Bayesian density estimation with location-scale mixtures
- Adaptive Bayesian estimation using a Gaussian random field with inverse gamma bandwidth
- Adaptive Bayesian inference on the mean of an infinite-dimensional normal distribution
- Adaptive Bayesian multivariate density estimation with Dirichlet mixtures
- Adaptive Bayesian nonparametric regression using a kernel mixture of polynomials with application to partial linear models
- Adaptive Bayesian procedures using random series priors
- Adaptive estimation of multivariate functions using conditionally Gaussian tensor-product spline priors
- Adaptive nonparametric Bayesian inference using location-scale mixture priors
- Asymptotic Statistics
- Bayesian estimation of sparse signals with a continuous spike-and-slab prior
- Bayesian optimal adaptive estimation using a sieve prior
- Bayesian regression tree ensembles that adapt to smoothness and sparsity
- Convergence rates for Bayesian density estimation of infinite-dimensional exponential families
- Convergence rates for posterior distributions and adaptive estimation
- Convergence rates of posterior distributions for non iid observations
- Convergence rates of posterior distributions.
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Entropies and rates of convergence for maximum likelihood and Bayes estimation for mixtures of normal densities.
- Fundamentals of nonparametric Bayesian inference
- Gaussian processes for machine learning.
- Generalized additive models
- High-dimensional additive modeling
- scientific article; zbMATH DE number 5190601 (Why is no real title available?)
- scientific article; zbMATH DE number 2001584 (Why is no real title available?)
- Information rates of nonparametric Gaussian process methods
- Mathematical foundations of infinite-dimensional statistical models
- Minimax-optimal nonparametric regression in high dimensions
- Minimax-optimal rates for sparse additive models over kernel classes via convex programming
- Needles and straw in a haystack: posterior concentration for possibly sparse sequences
- On adaptive posterior concentration rates
- On posterior consistency in nonparametric regression problems
- Optimal Bayesian estimation in random covariate design with a rescaled Gaussian process prior
- Optimal global rates of convergence for nonparametric regression
- Posterior asymptotics of nonparametric location-scale mixtures for multivariate density estimation
- Posterior concentration for Bayesian regression trees and forests
- Posterior concentration rates for infinite dimensional exponential families
- Posterior convergence rates of Dirichlet mixtures at smooth densities
- Posterior rates of convergence for Dirichlet mixtures of exponential power densities
- Rate exact Bayesian adaptation with modified block priors
- Rates of contraction of posterior distributions based on Gaussian process priors
- Rates of contraction with respect to \(L_2\)-distance for Bayesian nonparametric regression
- Sparse additive models
- Sparsity in multiple kernel learning
- Supremum norm posterior contraction and credible sets for nonparametric multivariate regression
- The spike-and-slab LASSO
- Wavelets on the interval and fast wavelet transforms
Cited in
(3)
This page was built for publication: Rates of contraction with respect to \(L_2\)-distance for Bayesian nonparametric regression
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2326065)
