Gaussian Mean Field Regularizes by Limiting Learned Information

Variational inference with a factorized Gaussian posterior estimate is a widely-used approach for learning parameters and hidden variables. Empirically, a regularizing effect can be observed that is poorly understood. In this work, we show how mean field inference improves generalization by limiting...

Descripción completa

Detalles Bibliográficos
Autores principales: Kunze, Julius, Kirsch, Louis, Ritter, Hippolyt, Barber, David
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2019
Materias:
Article
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7515287/
https://www.ncbi.nlm.nih.gov/pubmed/33267472
http://dx.doi.org/10.3390/e21080758

Internet

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7515287/
https://www.ncbi.nlm.nih.gov/pubmed/33267472
http://dx.doi.org/10.3390/e21080758

Ejemplares similares