Extending Stan for Deep Probabilistic Programming - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Preprints, Working Papers, ... Year : 2018

Extending Stan for Deep Probabilistic Programming

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.

Dates and versions

hal-03117782 , version 1 (21-01-2021)

Identifiers

Cite

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

Altmetric

Share

Gmail Facebook X LinkedIn More