Reproduce “A Compilation Target for Probabilistic Programming Languages” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Brooks Paige, Frank Wood
The authors introduce probabilistic C, a C-language intermediate representation for probabilistic inference that uses standard compilers and operating-system tools. Its particle-based methods use process forking to branch executions and resample them. In tests on two models, compiled particle Gibbs reached posterior estimates about 100 times faster than an interpreted implementation; this comparison reflects execution architectures, not compilation effects alone.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.