Logo image
Bayesian Implementation of the Factor‐Analytic Mixed Model and Application to Embeddings
Article de revue scientifique   Open Access   Avec comité de lecture

Bayesian Implementation of the Factor‐Analytic Mixed Model and Application to Embeddings

Alexandre Marchal, Maxence Lamure, Tanguy Lefort et Frank Imbach
Stat, Vol.15(3)
03/07/2026

Résumé

Bayesian methods genetic statistics Markov chain Monte Carlo multivariate analysis neural networks prediction
Mixed models and neural networks each offer complementary frameworks for prediction. Theoretically grounded in inference, mixed models can unveil latent variance structure notably with the factor‐analytic approach. In this article, we propose to bridge the gap between the two frameworks, leveraging embedding data from a neural network encoder to structure the covariance matrix in a mixed model. First, we implemented a Bayesian factor‐analytic mixed model and challenged it using simulations. The model showed sensitivity to the prior, with a propensity to overfitting under some hyperparameterizations. However, with an appropriate hyperparameterization it showed high predictive performance and a good ability to infer the latent structure (accuracy between 0.881 and 0.919 on the two first factorial axes). Using a challenging design in which each individual was observed a single time in a single environment, we showed that the embedding information enabled prediction of unobserved individuals. A small but significant ( p ‐value < 0.05) amount of real bibliometric data variance was captured using this factor‐analytic component applied within a negative‐binomial mixed model. These results call for further investigation of the use of the mixed model framework to analyse embedding data.

Fichiers et liens (2)

url
Find in HALAfficher
url
https://doi.org/10.1002/sta4.70164Afficher
Publié (version de la notice) Ouvrir

Indicateurs

1 Consultations de la notice

Détails

Logo image