paper-with-me

홈 › Papers

Learning Proposals for Probabilistic Programs with Inference Combinators

2021-03-01 · Sam Stites, Heiko Zimmermann, Hao Wu, Eli Sennesh, Jan-Willem van de Meent

We develop operators for construction of proposals in probabilistic programs, which we refer to as inference combinators. Inference combinators define a grammar over importance samplers that compose primitive operations such as application of a transition kernel and importance resampling. Proposals in these samplers can be parameterized using neural networks, which in turn can be trained by optimizing variational objectives. The result is a framework for user-programmable variational methods that are correct by construction and can be tailored to specific models. We demonstrate the flexibility of this framework by implementing advanced variational methods based on amortized Gibbs sampling and annealing.

📄 PDF Abstract BibTeX arXiv:2103.00668

Code (1)

probtorch/combinators 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Composing Modeling and Inference Operations with Probabilistic Program Combinators

2018-11-14 · Eli Sennesh, Adam Ścibior, Hao Wu, Jan-Willem van de Meent

Probabilistic programs with dynamic computation graphs can define measures over sample spaces with unbounded dimensionality, which constitute programmatic analogues to Bayesian nonparametrics. Owing to the generality of …

Using probabilistic programs as proposals

2018-01-11 · Marco F. Cusumano-Towner, Vikash K. Mansinghka

Monte Carlo inference has asymptotic guarantees, but can be slow when using generic proposals. Handcrafted proposals that rely on user knowledge about the posterior distribution can be efficient, but are difficult to der…

Probabilistic Programming

Applications of Probabilistic Programming (Master's thesis, 2015)

2016-05-31 · Yura N. Perov

This thesis describes work on two applications of probabilistic programming: the learning of probabilistic program code given specifications, in particular program code of one-dimensional samplers; and the facilitation o…

Object RecognitionProbabilistic Programming

Measure Transformer Semantics for Bayesian Machine Learning

2013-08-03 · Johannes Borgström, Andrew D. Gordon, Michael Greenberg, James Margetson 외

The Bayesian approach to machine learning amounts to computing posterior distributions of random variables from a probabilistic model of how the variables are related (that is, a prior distribution) and a set of observat…

BIG-bench Machine Learning

Programs as Polypeptides

2015-06-04 · Lance R. Williams

We describe a visual programming language for defining behaviors manifested by reified actors in a 2D virtual world that can be compiled into programs comprised of sequences of combinators that are themselves reified as …