Embedded-model flows: Combining the inductive biases of model-free deep learning and explicit probabilistic modeling

Normalizing flows have shown great success as general-purpose density estimators. However, many real world applications require the use of domain-specific knowledge, which normalizing flows cannot readily incorporate. We propose embedded-model flows (EMF), which alternate general-purpose transformat...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2022-03
Hauptverfasser:	Silvestri, Gianluigi, Fertig, Emily, Moore, Dave, Ambrogioni, Luca
Format:	Artikel
Sprache:	eng
Schlagworte:	Inference Probabilistic models Transformations (mathematics)
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!