Uncertainty quantification for sparse spectral variational approximations in Gaussian process regression
From MaRDI portal
Publication:6184877
Abstract: We investigate the frequentist properties of the variational sparse Gaussian Process regression model. In the theoretical analysis we focus on the variational approach with spectral features as inducing variables. We derive guarantees and limitations for the frequentist coverage of the resulting variational credible sets. We also derive sufficient and necessary lower bounds for the number of inducing variables required to achieve minimax posterior contraction rates. The implications of these results are demonstrated for different choices of priors. In a numerical analysis we consider a wider range of inducing variable methods and observe similar phenomena beyond the scope of our theoretical findings.
Cites work
- scientific article; zbMATH DE number 3002670 (Why is no real title available?)
- A Correspondence Between Bayesian Estimation on Stochastic Processes and Smoothing by Splines
- Adaptive Bayesian inference in the Gaussian sequence model using exponential-variance priors
- An analysis of Bayesian inference for nonparametric regression
- Asymptotic Statistics
- Asymptotic frequentist coverage properties of Bayesian credible sets for sieve priors
- Asymptotic methods in statistical decision theory
- Bayesian inference with rescaled Gaussian process priors
- Bayesian inverse problems with Gaussian priors
- Can we trust Bayesian uncertainty quantification from Gaussian process priors with squared exponential covariance kernel?
- Concentration of tempered posteriors and of their variational approximations
- Convergence of sparse variational inference in Gaussian processes regression
- Convergence rates of variational posterior distributions
- Frequentist consistency of variational Bayes
- Frequentist coverage of adaptive nonparametric Bayesian credible sets
- Gaussian processes for machine learning.
- Information-Theoretic Regret Bounds for Gaussian Process Optimization in the Bandit Setting
- Mercer's theorem on general domains: on the interaction between measures, kernels, and RKHSs
- On the Bernstein-von Mises phenomenon for nonparametric Bayes procedures
- Pattern recognition and machine learning.
- Rates of contraction of posterior distributions based on Gaussian process priors
- Variational Bayes for High-Dimensional Linear Regression With Sparse Priors
- \(\alpha\)-variational inference with statistical guarantees
Cited in
(2)
This page was built for publication: Uncertainty quantification for sparse spectral variational approximations in Gaussian process regression
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6184877)
