Gaussian variational approximation with sparse precision matrices
From MaRDI portal
Publication:1702005
DOI10.1007/S11222-017-9729-7zbMATH Open1384.62105arXiv1605.05622OpenAlexW3103982962MaRDI QIDQ1702005FDOQ1702005
David J. Nott, Linda S. L. Tan
Publication date: 27 February 2018
Published in: Statistics and Computing (Search for Journal in Brave)
Abstract: We consider the problem of learning a Gaussian variational approximation to the posterior distribution for a high-dimensional parameter, where we impose sparsity in the precision matrix to reflect appropriate conditional independence structure in the model. Incorporating sparsity in the precision matrix allows the Gaussian variational distribution to be both flexible and parsimonious, and the sparsity is achieved through parameterization in terms of the Cholesky factor. Efficient stochastic gradient methods which make appropriate use of gradient information for the target distribution are developed for the optimization. We consider alternative estimators of the stochastic gradients which have lower variation and are more stable. Our approach is illustrated using generalized linear mixed models and state space models for time series.
Full work available at URL: https://arxiv.org/abs/1605.05622
Recommendations
- Conditionally structured variational Gaussian approximation with importance weights
- The Variational Gaussian Approximation Revisited
- The limited-memory recursive variational Gaussian approximation (L-RVGA)
- Bayesian structure learning in graphical models
- Asymptotic normality and valid inference for Gaussian variational approximation
variational Bayesstochastic gradient algorithmsGaussian variational approximationsparse precision matrix
Bayesian inference (62F15) Learning and adaptive systems in artificial intelligence (68T05) Generalized linear models (logistic models) (62J12)
Cites Work
- Title not available (Why is that?)
- Title not available (Why is that?)
- Semiparametric mean field variational Bayes: general principles and numerical issues
- A new approach to Cholesky-based covariance regularization in high dimensions
- Approximate Bayesian inference for latent Gaussian models by using integrated nested Laplace approximations (with discussion)
- Stochastic Volatility: Likelihood Inference and Comparison with ARCH Models
- Approximate Inference in Generalized Linear Mixed Models
- Title not available (Why is that?)
- Title not available (Why is that?)
- Explaining Variational Approximations
- A Stochastic Approximation Method
- Title not available (Why is that?)
- Some Covariance Models for Longitudinal Count Data with Overdispersion
- Maximizing Generalized Linear Mixed Model Likelihoods With an Automated Monte Carlo EM Algorithm
- Multivariate Stochastic Variance Models
- Introduction to Stochastic Search and Optimization
- Variational inference for generalized linear mixed models using partially noncentered parametrizations
- Title not available (Why is that?)
- Approximate Dynamic Programming
- A stochastic variational framework for fitting and diagnosing generalized linear mixed models
- Time series analysis by state space methods.
- An introduction to variational methods for graphical models
- Applied Logistic Regression
- The Variational Gaussian Approximation Revisited
- Fixed-form variational posterior approximation through stochastic linear regression
- Streamlined mean field variational Bayes for longitudinal and multilevel data analysis
Cited In (24)
- Conditionally structured variational Gaussian approximation with importance weights
- Bayesian computational methods for state-space models with application to SIR model
- Variational Bayes with synthetic likelihood
- Variational inference for high dimensional structured factor copulas
- The Variational Gaussian Approximation Revisited
- Variational Bayes for Fast and Accurate Empirical Likelihood Inference
- R-VGAL: a sequential variational Bayes algorithm for generalised linear mixed models
- Generalized Sparse Precision Matrix Selection for Fitting Multivariate Gaussian Random Fields to Large Data Sets
- Bayesian Deep Net GLM and GLMM
- An integrated FIR adaptive filter design by hybridizing canonical signed digit (CSD) and approximate booth recode (ABR) algorithm in DA architecture for the reduction of noise in the sensor nodes
- Variational approximation for heteroscedastic linear models and matching pursuit algorithms
- Stochastic variational inference for GARCH models
- On coregionalized multivariate Gaussian Markov random fields: construction, parameterization, and Bayesian estimation and inference
- Gaussian Variational Approximation With a Factor Covariance Structure
- The limited-memory recursive variational Gaussian approximation (L-RVGA)
- Updating variational Bayes: fast sequential posterior inference
- Implementation and analysis of GPU algorithms for Vecchia approximation
- Deep Compositional Spatial Models
- Bayesian Variational Inference for Exponential Random Graph Models
- High-Dimensional Copula Variational Approximation Through Transformation
- Gaussian variational approximations for high-dimensional state space models
- Variational Bayes Estimation of Discrete-Margined Copula Models With Application to Time Series
- Variational Bayesian approximation of inverse problems using sparse precision matrices
- A variational inference for the Lévy adaptive regression with multiple kernels
Uses Software
This page was built for publication: Gaussian variational approximation with sparse precision matrices
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1702005)
