Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Composing inference algorithms as program transformations
(Submitted on 6 Mar 2016 (v1), last revised 12 Jul 2017 (this version, v2))
Abstract: Probabilistic inference procedures are usually coded painstakingly from scratch, for each target model and each inference algorithm. We reduce this effort by generating inference procedures from models automatically. We make this code generation modular by decomposing inference algorithms into reusable program-to-program transformations. These transformations perform exact inference as well as generate probabilistic programs that compute expectations, densities, and MCMC samples. The resulting inference procedures are about as accurate and fast as other probabilistic programming systems on real-world problems.
Submission history
From: Robert Zinkov [view email][v1] Sun, 6 Mar 2016 21:30:10 GMT (27kb)
[v2] Wed, 12 Jul 2017 16:01:42 GMT (273kb,D)
Link back to: arXiv, form interface, contact.