Transforming Probabilistic Programs into Algebraic Circuits for Inference and Learning

Abstract

Probabilistic (logic) programs are routinely compiled into arithmetic circuits. During such a compilation step, the logic representation of a probabilistic program is transformed into an arithmetic representation. We show that this transformation and the resulting circuits cannot only be used for discrete probabilistic inference, but also for a number of other tasks such as differentiation, learning and probabilistic inference in the discrete-continuous domain.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

Dos Martires et al. "Transforming Probabilistic Programs into Algebraic Circuits for Inference and Learning." NeurIPS 2019 Workshops: Program_Transformations, 2019.

Markdown

[Dos Martires et al. "Transforming Probabilistic Programs into Algebraic Circuits for Inference and Learning." NeurIPS 2019 Workshops: Program_Transformations, 2019.](https://mlanthology.org/neuripsw/2019/martires2019neuripsw-transforming/)

BibTeX

@inproceedings{martires2019neuripsw-transforming,
  title     = {{Transforming Probabilistic Programs into Algebraic Circuits for Inference and Learning}},
  author    = {Dos Martires, Pedro Zuidberg and Derkinderen, Vincent and Manhaeve, Robin and Meert, Wannes and Kimmig, Angelika and De Raedt, Luc},
  booktitle = {NeurIPS 2019 Workshops: Program_Transformations},
  year      = {2019},
  url       = {https://mlanthology.org/neuripsw/2019/martires2019neuripsw-transforming/}
}