Compiling Stan to Generative Probabilistic Languages and Extension to Deep Probabilistic Programming
This provides incremental improvements for statisticians and users of Stan by enhancing performance and adding features without requiring them to learn new languages.
The paper tackled the problem of compiling Stan models to generative probabilistic languages, enabling new backends for Pyro and NumPyro, resulting in a 2.3x speedup with NumPyro over Stan on benchmarks and extending Stan with variational inference and deep models.
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 2.3x speedup compared to Stan in geometric mean over 26 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.