Matlab code implementing Hamiltonian Annealed Importance Sampling for importance weight, partition function, and log likelihood estimation for models with continuous state spaces
Code to perform Hamiltonian Annealed Importance Sampling for log
likelihood estimation, partition function estimation, and importance
weight estimation in models with intractable partition functions. Can
also be used for Hamiltonian Monte Carlo sampling (single step, with
partial momentum refreshment).
HAIS_examples.m demonstrates the capabilities of this code in a
variety of scenarios.
HAIS.m performs Hamiltonian Annealed Importance Sampling.
HAIS_logL.m calculates the log likelihood of a model given data using
HAIS.
HAIS_logL_aux.m calculates the log likelihood of a model with hidden
(auxiliary) variables given data using HAIS.
See included PDF HAIS.pdf for a description of the Hamiltonian
Annealed Importance technique:
J Sohl-Dickstein, BJ Culpepper. Hamiltonian annealed importance sampling
for partition function estimation. Redwood Technical Report. 2011.