Laplace approximation for logistic Gaussian process density estimation and regression
From MaRDI portal
Publication:899031
DOI10.1214/14-BA872zbMATH Open1327.62248arXiv1211.0174WikidataQ61703745 ScholiaQ61703745MaRDI QIDQ899031FDOQ899031
Publication date: 21 December 2015
Published in: Bayesian Analysis (Search for Journal in Brave)
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.
Full work available at URL: https://arxiv.org/abs/1211.0174
density estimationGaussian processLaplace's methoddensity regressionapproximate inferencelogistic transformation
Bayesian inference (62F15) Density estimation (62G07) Nonparametric regression and quantile regression (62G08)
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- Bayesian Model Assessment and Comparison Using Cross-Validation Predictive Densities
- Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations (with discussion)
- A Finite Element Method for Density Estimation with Gaussian Process Priors
- Towards a practicable Bayesian nonparametric density estimator
- Sequential Imputations and Bayesian Missing Data Problems
- Prior distributions for variance parameters in hierarchical models (Comment on article by Browne and Draper)
- Adaptive Bayesian estimation using a Gaussian random field with inverse gamma bandwidth
- Bayesian Inference in Econometric Models Using Monte Carlo Integration
- Default priors for density estimation with mixture models
- Posterior consistency of logistic Gaussian process priors in density estimation
- Bayesian density regression with logistic Gaussian process and subspace projection
- Latent factor models for density estimation
- A Bayesian Approach to Density Estimation
- The Multivariate Split Normal Distribution and Asymmetric Principal Components Analysis
Cited In (13)
- A Geometric Variational Approach to Bayesian Inference
- Parallel Gaussian process surrogate Bayesian inference with noisy likelihood evaluations
- A randomized multi-index sequential Monte Carlo method
- Flexible Mixture-Amount Models Using Multivariate Gaussian Processes
- Adaptive Conditional Distribution Estimation with Bayesian Decision Tree Ensembles
- A review of uncertainty quantification for density estimation
- Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC
- Robust importance sampling with adaptive winsorization
- Title not available (Why is that?)
- Title not available (Why is that?)
- Title not available (Why is that?)
- A scalable Matérn Gaussian process for learning spatial curves distributions
- Approximate lognormality of the sample semi-variogram under a gaussian process
Uses Software
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)
