Edward: A library for probabilistic modeling, inference, and criticism

10/31/2016
by   Dustin Tran, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
03/02/2016

Automatic Differentiation Variational Inference

Probabilistic modeling is iterative. A scientist posits a simple model, ...
research
08/09/2019

Probabilistic Models with Deep Neural Networks

Recent advances in statistical inference have significantly expanded the...
research
01/13/2017

Deep Probabilistic Programming

We propose Edward, a Turing-complete probabilistic programming language....
research
11/30/2018

TF-Ranking: Scalable TensorFlow Library for Learning-to-Rank

TensorFlow Ranking is the first open source library for solving large-sc...
research
01/11/2019

SPFlow: An Easy and Extensible Library for Deep Probabilistic Learning using Sum-Product Networks

We introduce SPFlow, an open-source Python library providing a simple in...
research
07/01/2016

Probabilistic Programming and PyMC3

In recent years sports analytics has gotten more and more popular. We pr...
research
03/22/2019

A Model Counter's Guide to Probabilistic Systems

In this paper, we systematize the modeling of probabilistic systems for ...

Please sign up or login with your details

Forgot password? Click here to reset