Probabilistic modeling is a powerful approach for analyzing empirical
information. We describe Edward, a library for probabilistic modeling. Edward's
design reflects an iterative process pioneered by George Box: build a model of
a phenomenon, make inferences about the model given data, and criticize the
model's fit to the data. Edward supports a broad class of probabilistic models,
efficient algorithms for inference, and many techniques for model criticism.
The library builds on top of TensorFlow to support distributed training and
hardware such as GPUs. Edward enables the development of complex probabilistic
models and their algorithms at a massive scale.
%0 Generic
%1 tran2017edward
%A Tran, Dustin
%A Kucukelbir, Alp
%A Dieng, Adji B.
%A Rudolph, Maja
%A Liang, Dawen
%A Blei, David M.
%D 2017
%K bayesian advi deep_learning expectation_maximization machine_learning mcmc optimization probabilistic_programming variational_inference
%T Edward: A library for probabilistic modeling, inference, and criticism
%U http://arxiv.org/abs/1610.09787
%X Probabilistic modeling is a powerful approach for analyzing empirical
information. We describe Edward, a library for probabilistic modeling. Edward's
design reflects an iterative process pioneered by George Box: build a model of
a phenomenon, make inferences about the model given data, and criticize the
model's fit to the data. Edward supports a broad class of probabilistic models,
efficient algorithms for inference, and many techniques for model criticism.
The library builds on top of TensorFlow to support distributed training and
hardware such as GPUs. Edward enables the development of complex probabilistic
models and their algorithms at a massive scale.
@misc{tran2017edward,
abstract = {{Probabilistic modeling is a powerful approach for analyzing empirical
information. We describe Edward, a library for probabilistic modeling. Edward's
design reflects an iterative process pioneered by George Box: build a model of
a phenomenon, make inferences about the model given data, and criticize the
model's fit to the data. Edward supports a broad class of probabilistic models,
efficient algorithms for inference, and many techniques for model criticism.
The library builds on top of TensorFlow to support distributed training and
hardware such as GPUs. Edward enables the development of complex probabilistic
models and their algorithms at a massive scale.}},
added-at = {2018-12-07T09:10:16.000+0100},
archiveprefix = {arXiv},
author = {Tran, Dustin and Kucukelbir, Alp and Dieng, Adji B. and Rudolph, Maja and Liang, Dawen and Blei, David M.},
biburl = {https://www.bibsonomy.org/bibtex/2cadedb225ee76c51abba517d8c3059bd/jpvaldes},
citeulike-article-id = {14178239},
citeulike-linkout-0 = {http://arxiv.org/abs/1610.09787},
citeulike-linkout-1 = {http://arxiv.org/pdf/1610.09787},
day = 1,
eprint = {1610.09787},
interhash = {a2006aef19188a7a604d361e3eac94fb},
intrahash = {cadedb225ee76c51abba517d8c3059bd},
keywords = {bayesian advi deep_learning expectation_maximization machine_learning mcmc optimization probabilistic_programming variational_inference},
month = feb,
posted-at = {2017-05-26 13:14:03},
priority = {4},
timestamp = {2018-12-07T09:41:54.000+0100},
title = {{Edward: A library for probabilistic modeling, inference, and criticism}},
url = {http://arxiv.org/abs/1610.09787},
year = 2017
}