Sampling

This module contains implementations of various SGMCMC samplers which are well-suited for Bayesian Deep Learning. For instance, all samplers implemented in this module can be easily used to learn a BayesianNeuralNetwork. .. XXX: Cross reference to BNN doku (and to samplers?)

Samplers:

• Base Classes for MCMC Methods
• Stochastic Gradient Hamiltonian Monte Carlo (SGHMC)
• Stochastic Gradient Langevin Dynamics (SGLD)
• Relativistic Stochastic Gradient Hamiltonian Monte Carlo (RelSGHMC)