Reference : Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Sc...
E-prints/Working papers : Already available on another site
Engineering, computing & technology : Computer science
Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators
Lezcano Casado, Mario [> >]
Gunes Baydin, Atilim [> >]
Martinez Rubio, David [> >]
Le, Tuan Anh [> >]
Wood, Frank [> >]
Heinrich, Lukas [> >]
Louppe, Gilles mailto [Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data >]
Cranmer, Kyle [> >]
Ng, Karen [> >]
Bhimji, Wahid [> >]
Prabhat [> >]
[en] Computer Science - Artificial Intelligence ; Physics - Data Analysis ; Statistics and Probability ; 68T37 ; 68T05 ; 62P35 ; G.3 ; I.2.6 ; J.2
[en] We consider the problem of Bayesian inference in the family of probabilistic models implicitly defined by stochastic generative models of data. In scientific fields ranging from population biology to cosmology, low-level mechanistic components are composed to create complex generative models. These models lead to intractable likelihoods and are typically non-differentiable, which poses challenges for traditional approaches to inference. We extend previous work in "inference compilation", which combines universal probabilistic programming and deep learning methods, to large-scale scientific simulators, and introduce a C++ based probabilistic programming library called CPProb. We successfully use CPProb to interface with SHERPA, a large code-base used in particle physics. Here we describe the technical innovations realized and planned for this library.
7 pages, 2 figures

File(s) associated to this reference

Fulltext file(s):

Open access
1712.07901.pdfAuthor preprint2.3 MBView/Open

Bookmark and Share SFX Query

All documents in ORBi are protected by a user license.