blei-lab / ars-reparameterization

Source code for Naesseth et. al. "Reparameterization Gradients through Acceptance-Rejection Sampling Algorithms" (2017)

ars-reparameterization

This Git-repository includes source code to run experiments and generate figures for

Reparameterization Gradients through Acceptance-Rejection Sampling Algorithms.
Christian A. Naesseth, Francisco J. R. Ruiz, Scott W. Linderman, and David M. Blei
Proceedings of the 20th International Conference on Artificial Intelligence and Statistics 2017,
Fort Lauderdale, Florida, USA.

other

Jupyter notebook to generate Figure 1 in the paper.

gamma

Jupyter notebooks to generate Figure 2 and 3.

dirichlet

Jupyter notebooks and Python code to run experiments for the Dirichlet-Multionomial example, as well as generating Figure 4.

sparse gamma def

Jupyter notebooks and Python code to run experiments for the Sparse Gamma DEF example. Also includes results and data used to generate Figure 5 and Table 1.