Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Extending Stan for Deep Probabilistic Programming

Guillaume Baudart 1 Javier Burroni 2 Martin Hirzel 3 Louis Mandel 3 Avraham Shinnar 3
1 Parkas - Parallélisme de Kahn Synchrone
Inria de Paris, DI-ENS - Département d'informatique - ENS Paris, CNRS - Centre National de la Recherche Scientifique
Abstract : Stan is a probabilistic programming language that is popular in the statistics community, with a high-level syntax for expressing probabilistic models. Stan differs by nature from generative probabilistic programming languages like Church, Anglican, or Pyro. This paper presents a comprehensive compilation scheme to compile any Stan model to a generative language and proves its correctness. We use our compilation scheme to build two new backends for the Stanc3 compiler targeting Pyro and NumPyro. Experimental results show that the NumPyro backend yields a 4.1x speedup compared to Stan in geometric mean over 25 benchmarks. Building on Pyro we extend Stan with support for explicit variational inference guides and deep probabilistic models. That way, users familiar with Stan get access to new features without having to learn a fundamentally new language.
Document type :
Preprints, Working Papers, ...
Complete list of metadata
Contributor : Guillaume Baudart Connect in order to contact the contributor
Submitted on : Thursday, January 21, 2021 - 3:16:14 PM
Last modification on : Friday, January 21, 2022 - 3:19:05 AM

Links full text


  • HAL Id : hal-03117782, version 1
  • ARXIV : 1810.00873



Guillaume Baudart, Javier Burroni, Martin Hirzel, Louis Mandel, Avraham Shinnar. Extending Stan for Deep Probabilistic Programming. 2018. ⟨hal-03117782⟩



Les métriques sont temporairement indisponibles