Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 13, 2025Open Access

Scalable Variable Selection and Model Averaging for Latent Regression Models Using Approximate Variational Bayes

View Full Paper
Ask AI
Bookmark
Share

Authors

GZGregor ZensInternational Institute for Applied Systems AnalysisMSMark F. J. SteelUniversity of Warwick

Discussion

Loading...

Member takes

Implication

A novel method addresses computational costs and marginal likelihoods in Bayesian model averaging for latent regression, highlighting model selection consistency.

Key Points

  • The proposed method demonstrates asymptotic model selection consistency, preserving accuracy while reducing computation time significantly.
  • By using an approximate variational Bayes scheme, the approach addresses large model spaces effectively with reduced runtime.
  • Numerical studies across various models, including probit and Poisson log-normal regression, confirm accuracy and substantial speed improvements.
  • Practical applications reveal the method's effectiveness for Bayesian inference in large datasets and under model uncertainties.

Cite This Study

Zens et al. (2025) studied this question.

synapsesocial.com/papers/68ecfebf950606aabec095c1https://doi.org/10.48550/arxiv.2509.11751
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Variable selection for semivarying coefficient models via local averaging2024
  2. 2Bayesian model selection via mean-field variational approximation2024 · 11 citations
  3. 3Estimating Discrete Latent Variable Models Using Amortized Variational Inference2026
  4. 4Mean-field variational Bayes for sparse probit regression2026
  5. 5Variational Inference for Latent Variable Models in High Dimensions2025