Hamiltonian Monte Carlo


Hamiltonian Monte Carlo (HMC) is an MCMC algorithm which makes use of gradient information in order to avoid random walks and move more quickly toward regions of high probability. It is based on a discretization of Hamiltonian dynamics, with a Metropolis-Hastings accept/reject step to ensure that it has the right stationary distribution.


This concept has the prerequisites:

Core resources (we're sorry, we haven't finished tracking down resources for this concept yet)

Supplemental resources (the following are optional, but you may find them useful)


Bayesian Reasoning and Machine Learning
A textbook for a graudate machine learning course.
Author: David Barber
Machine learning summer school: Markov chain Monte Carlo (2009)
A video tutorial on MCMC methods.
Location: Part 2, 22:58 to 24:10 and 39:44 to 43:53
Author: Iain Murray
Information Theory, Inference, and Learning Algorithms
A graudate-level textbook on machine learning and information theory.
Author: David MacKay


See also