RECORD
Variational Inference for Generalized Linear Mixed Models Using Partially Noncentered Parametrizations
Computation · Submitted 17 May 2012
abstract compiled · 2026-09-23 · no language-model review
The authors' main claim Third, we show that partial noncentering can accelerate convergence and produce more accurate posterior approximations than centering or noncentering.
The compiled record
The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
Open the compiled record
The abstract, sentence by sentence
The effects of different parametrizations on the convergence of Bayesian computational algorithms for hierarchical models are well explored. Techniques such as centering, noncentering and partial noncentering can be used to accelerate convergence in MCMC and EM algorithms but are still not well studied for variational Bayes (VB) methods. As a fast deterministic approach to posterior approximation, VB is attracting increasing interest due to its suitability for large high-dimensional data. Use of different parametrizations for VB has not only computational but also statistical implications, as different parametrizations are associated with different factorized posterior approximations. We examine the use of partially noncentered parametrizations in VB for generalized linear mixed models (GLMMs). Our paper makes four contributions. First, we show how to implement an algorithm called nonconjugate variational message passing for GLMMs. Second, we show that the partially noncentered parametrization can adapt to the quantity of information in the data and determine a parametrization close to optimal. Third, we show that partial noncentering can accelerate convergence and produce more accurate posterior approximations than centering or noncentering. Finally, we demonstrate how the variational lower bound, produced as part of the computation, can be useful for model selection.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | The effects of different parametrizations on the convergence of Bayesian computational algorithms for hierarchical models are well explored. | background 1.00 | 0.04 | 0.02 | asserted 1.00 |
| s1 | Techniques such as centering, noncentering and partial noncentering can be used to accelerate convergence in MCMC and EM algorithms but are still not well studied for variational Bayes (VB) methods. | gap 0.94 | 0.04 | 0.04 | asserted 0.93 |
| s2 | As a fast deterministic approach to posterior approximation, VB is attracting increasing interest due to its suitability for large high-dimensional data. | background 1.00 | 0.04 | 0.02 | asserted 0.99 |
| s3 | Use of different parametrizations for VB has not only computational but also statistical implications, as different parametrizations are associated with different factorized posterior approximations. | background 0.91 | 0.06 | 0.04 | asserted 0.99 |
| s4 | We examine the use of partially noncentered parametrizations in VB for generalized linear mixed models (GLMMs). | contribution 0.97 | 0.73 | 0.05 | asserted 0.73 |
| s5 | Our paper makes four contributions. | contribution 0.99 | 0.38 | 0.03 | asserted 0.89 |
| s6 | First, we show how to implement an algorithm called nonconjugate variational message passing for GLMMs. | contribution 0.99 | 0.95 | 0.06 | asserted 1.00 |
| s7 | Second, we show that the partially noncentered parametrization can adapt to the quantity of information in the data and determine a parametrization close to optimal. | contribution 0.62 | 0.93 | 0.27 | asserted 0.99 |
| s8 | Third, we show that partial noncentering can accelerate convergence and produce more accurate posterior approximations than centering or noncentering. | result 0.56 | 0.94 | 0.52 | asserted 1.00 |
| s9 | Finally, we demonstrate how the variational lower bound, produced as part of the computation, can be useful for model selection. | contribution 0.99 | 0.94 | 0.11 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v3
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
The paper
Original abstract
The effects of different parametrizations on the convergence of Bayesian computational algorithms for hierarchical models are well explored. Techniques such as centering, noncentering and partial noncentering can be used to accelerate convergence in MCMC and EM algorithms but are still not well studied for variational Bayes (VB) methods. As a fast deterministic approach to posterior approximation, VB is attracting increasing interest due to its suitability for large high-dimensional data. Use of different parametrizations for VB has not only computational but also statistical implications, as different parametrizations are associated with different factorized posterior approximations. We examine the use of partially noncentered parametrizations in VB for generalized linear mixed models (GLMMs). Our paper makes four contributions. First, we show how to implement an algorithm called nonconjugate variational message passing for GLMMs. Second, we show that the partially noncentered parametrization can adapt to the quantity of information in the data and determine a parametrization close to optimal. Third, we show that partial noncentering can accelerate convergence and produce more accurate posterior approximations than centering or noncentering. Finally, we demonstrate how the variational lower bound, produced as part of the computation, can be useful for model selection.
Human review
How the literature uses it
What this paper does with the arXiv papers it cites
Discussion
Know something about this paper? Leave a comment, or highlight any passage above to comment on it.