Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming
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 3.7x speedup in geometric mean compared to Stan on 27 existing 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.
Fri 22 OctDisplayed time zone: Central Time (US & Canada) change
10:50 - 12:10
|Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming|
|On Probabilistic Termination of Functional Programs with Continuous Distributions|
|SPPL: Probabilistic Programming with Fast Exact Symbolic Inference|
Feras Saad Massachusetts Institute of Technology, Martin C. Rinard Massachusetts Institute of Technology, Vikash K. Mansinghka MITDOI
|Cyclic Program Synthesis|
Shachar Itzhaky Technion, Hila Peleg Technion, Nadia Polikarpova University of California at San Diego, Reuben N. S. Rowe University of Kent, Ilya Sergey National University of SingaporeDOI
|Discussion, Questions and Answers|