This folder contains examples related to Bayesian Methods.
We currently have Stochastic Gradient Langevin Dynamics (SGLD) (Welling and Teh, 2011) and Bayesian Dark Knowledge (BDK) (Balan, Rathod, Murphy and Welling, 2015).
sgld.ipynb shows how to use MXNet to repeat the toy experiment in the original SGLD paper.
bdk.ipynb shows how to use MXNet to implement the DistilledSGLD algorithm in Bayesian Dark Knowledge.
bdk_demo.py contains scripts (more than the notebook) related to Bayesian Dark Knowledge. Use python bdk_demo.py -d 1 -l 2 -t 50000 to run classification on MNIST.