In case you missed them, here are some articles from November 2017 of particular interest to users of Probabilistic Programming Languages(PPL).
Uber AI Lab open sourced their deep probabilistic programming language called Pyro. Pyro is a tool for deep probabilistic modeling, unifying the best of modern deeplearning and Bayesian modeling.
Prof. Aki Vehtari prepared this case study to demonstrate how to implement user defined probability functions in Stan.
“Statistical Rethinking” book by Richard McElreath was ported to PyMC3 and Python.
A message by Christopher Fonnesbeck on Theano and the Future of PyMC.
Join this discussion to help shape the future of PyMC3. With Theano development stopping, a new backend is needed for PyMC.
A study using PyMC3 about penalty scoring & saving ability in Major League Soccer (MLS). The conclusion was that it is hard to tell who is good at them.
In this video, Prof. Zoubin Ghahramani (Uber AI labs & University of Cambridge) made the case for the use of Probabilistic Programming.
Blei et al. (2017) prepared this gentle review of variational inference for non-ML researchers.
A lecture series to help you get started with Bayesian modeling using PyMC3.
Slides by Austin Rochford explaining NBA foul calls with Python and PyMC3.
A blog post by Jim Savage on Bayesian Instrumental variables in Stan.
A Blog post about diffusion/Wiener model analysis with Stan using R package brms.
Kevin Pei’s blog post on hierarchical Bayesian rating in PyMC3 with application to eSports.