Sciweavers

Share
FLOPS
2016
Springer

Probabilistic Inference by Program Transformation in Hakaru (System Description)

3 years 17 days ago
Probabilistic Inference by Program Transformation in Hakaru (System Description)
Abstract. We present Hakaru, a new probabilistic programming system that allows composable reuse of distributions, queries, and inference algorithms, all expressed in a single language of measures. The system implements two automatic and semantics-preserving program transformations—disintegration, which calculates conditional distributions, and simplification, which subsumes exact inference by computer algebra. We show how these features work together by describing the ideal workflow of a Hakaru user on two small problems. We highlight our composition of transformations and types in design and implementation.
Praveen Narayanan, Jacques Carette, Wren Romano, C
Added 03 Apr 2016
Updated 03 Apr 2016
Type Journal
Year 2016
Where FLOPS
Authors Praveen Narayanan, Jacques Carette, Wren Romano, Chung-chieh Shan, Robert Zinkov
Comments (0)
books