Laplace approximation for logistic Gaussian process density estimation and regression
From MaRDI portal
(Redirected from Publication:899031)
Abstract: Logistic Gaussian process (LGP) priors provide a flexible alternative for modelling unknown densities. The smoothness properties of the density estimates can be controlled through the prior covariance structure of the LGP, but the challenge is the analytically intractable inference. In this paper, we present approximate Bayesian inference for LGP density estimation in a grid using Laplace's method to integrate over the non-Gaussian posterior distribution of latent function values and to determine the covariance function parameters with type-II maximum a posteriori (MAP) estimation. We demonstrate that Laplace's method with MAP is sufficiently fast for practical interactive visualisation of 1D and 2D densities. Our experiments with simulated and real 1D data sets show that the estimation accuracy is close to a Markov chain Monte Carlo approximation and state-of-the-art hierarchical infinite Gaussian mixture models. We also construct a reduced-rank approximation to speed up the computations for dense 2D grids, and demonstrate density regression with the proposed Laplace approach.
Recommendations
- scientific article; zbMATH DE number 774847
- Bayesian density regression with logistic Gaussian process and subspace projection
- A finite element method for density estimation with Gaussian process priors
- On a generalization of the Laplace approximation
- Assessing approximate inference for binary Gaussian process classification.
Cites work
- scientific article; zbMATH DE number 3619167 (Why is no real title available?)
- scientific article; zbMATH DE number 1077338 (Why is no real title available?)
- scientific article; zbMATH DE number 1522714 (Why is no real title available?)
- scientific article; zbMATH DE number 6253886 (Why is no real title available?)
- A Bayesian Approach to Density Estimation
- A finite element method for density estimation with Gaussian process priors
- A unifying view of sparse approximate Gaussian process regression
- Adaptive Bayesian estimation using a Gaussian random field with inverse gamma bandwidth
- Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations (with discussion)
- Approximations for binary Gaussian process classification
- Bayesian Inference in Econometric Models Using Monte Carlo Integration
- Bayesian Model Assessment and Comparison Using Cross-Validation Predictive Densities
- Bayesian density regression with logistic Gaussian process and subspace projection
- Default priors for density estimation with mixture models
- GPstuff: Bayesian modeling with Gaussian processes
- Gaussian processes for machine learning.
- Latent factor models for density estimation
- Nested expectation propagation for Gaussian process classification with a multinomial probit likelihood
- Posterior consistency of logistic Gaussian process priors in density estimation
- Prior distributions for variance parameters in hierarchical models (Comment on article by Browne and Draper)
- Robust Gaussian process regression with a Student-\(t\) likelihood
- Sequential Imputations and Bayesian Missing Data Problems
- The Multivariate Split Normal Distribution and Asymmetric Principal Components Analysis
- Towards a practicable Bayesian nonparametric density estimator
Cited in
(15)- Flexible Mixture-Amount Models Using Multivariate Gaussian Processes
- Parallel Gaussian process surrogate Bayesian inference with noisy likelihood evaluations
- Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC
- A geometric variational approach to Bayesian inference
- Bayesian inference for spatio-temporal spike-and-slab priors
- scientific article; zbMATH DE number 7625160 (Why is no real title available?)
- Approximate lognormality of the sample semi-variogram under a gaussian process
- Adaptive Conditional Distribution Estimation with Bayesian Decision Tree Ensembles
- A randomized multi-index sequential Monte Carlo method
- Bayesian density regression with logistic Gaussian process and subspace projection
- Robust importance sampling with adaptive winsorization
- scientific article; zbMATH DE number 7626749 (Why is no real title available?)
- Laplace approximation and natural gradient for Gaussian process regression with heteroscedastic Student-\(t\) model
- A scalable Matérn Gaussian process for learning spatial curves distributions
- A review of uncertainty quantification for density estimation
This page was built for publication: Laplace approximation for logistic Gaussian process density estimation and regression
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q899031)
