Aki Vehtari

New paper "To select or not to select: predictively consistent priors instead of model selection" with Anna Elisabeth Riha, Leevi Lindgren, @davidkohns.bsky.social, @paulbuerkner.com arxiv.org/abs/2606.22850 Model selection is not a substitute for building good models in the f…

To select or not to select: predictively consistent priors instead of model selection arxiv.org

AI Weekly's analysis →

The paper claims flexible models with predictively consistent priors typically match or outperform selected simpler models in out-of-sample predictive performance.
When model selection does help, the authors argue it signals problematic prior specifications rather than a genuine modeling advantage.
The framework is tested across linear and logistic regression, variable selection, and nonlinear modeling, in a 34 page paper with 36 supplementary pages.

Read full analysis →

View on Bluesky · ♥ 101 ↻ 30 ↩ 1 · 2 from the directory shared this · 15d ago

↻ Aki Vehtari reposted

@statco-bot.bsky.social

Trevor Campbell, Jonathan H. Huggins, Kyurae Kim, Charles C. Margossian: Large-scale empirical tuning and comparison of default optimizers for variational inference https://arxiv.org/abs/2606.07841 https://arxiv.org/pdf/2606.07841 https://arxiv.org/html/2606.07841

Large-scale empirical tuning and comparison of default optimizers for variational inference arxiv.org View on Bluesky →

↻ Aki Vehtari reposted

@statco-bot.bsky.social

Trevor Campbell, Jonathan H. Huggins, Kyurae Kim, Charles C. Margossian: Large-scale empirical tuning and comparison of default optimizers for variational inference https://arxiv.org/abs/2606.07841 https://arxiv.org/pdf/2606.07841 https://arxiv.org/html/2606.07841

arxiv.org View on Bluesky →

↻ Aki Vehtari reposted

@statco-bot.bsky.social

Trevor Campbell, Jonathan H. Huggins, Kyurae Kim, Charles C. Margossian: Large-scale empirical tuning and comparison of default optimizers for variational inference https://arxiv.org/abs/2606.07841 https://arxiv.org/pdf/2606.07841 https://arxiv.org/html/2606.07841

Large-scale empirical tuning and comparison of default optimizers for variational inference arxiv.org View on Bluesky →

Some of the case studies are close to corresponding book chapters or have even more description, so check few more online. You can also read the arXiv version arxiv.org/abs/2011.01808 as a preview

Bayesian Workflow arxiv.org

View on Bluesky · ♥ 2 ↻ 1 ↩ 0 · 29d ago

↻ Aki Vehtari reposted

@paperposterbot.bsky.social

arXiv📈🤖 On Asymptotic Outlier Rejection in Bayesian Mixed Poisson Regression Models Under Extreme Target and Covariate Values By Pia, Vanhatalo

On Asymptotic Outlier Rejection in Bayesian Mixed Poisson Regression Models Under Extreme Target and Covariate Values arxiv.org View on Bluesky →

↻ Aki Vehtari reposted

@arxiv-stat-me.bsky.social

Miles Moran (Oregon State University), Rob Trangucci (Oregon State University), Lisa Madsen (Oregon State University) Bayesian Inference of Mixing and Transmission Heterogeneity in Stratified Disease Surveillance Models https://arxiv.org/abs/2605.29081

Bayesian Inference of Mixing and Transmission Heterogeneity in Stratified Disease Surveillance Models arxiv.org View on Bluesky →

Articles & links

In Aki Vehtari's orbit