Commit e35d77f1 authored by Christoph Lienhard's avatar Christoph Lienhard

Update to new paper abstract

parent cf985a5e
......@@ -4,24 +4,22 @@ HMCF - Hamiltonian Monte Carlo Sampling for Fields
HMCF implements a Hamiltonian Monte Carlo (HMC) sampler for the NIFTy
(“Numerical Information Field Theory”) framework.
It is available for Python3 on Unix-like systems.
Chances are that it also works with python 2.7, but this is not tested.
We guess real adventurers could even be so bold to try it on Windows using
Anaconda or equivalent frameworks.
HMCF offers an easy to use sampler if an appropriate NIFTy Energy is already
This can help estimating the impact of approximations present in other
approaches, or enable tackling entirely new problems.
It automatically adjusts the many free parameters steering the HMC sampling
machinery such as integration step size and the mass
matrix according to the requirements of field inference.
Furthermore, different convergence measures are available to check whether the
burn-in phase has finished.
Multiprocessing in the sense of running individual Markov chains (MC) on several
cores is possible as well.
HMCF “Hamiltonian Monte Carlo for Fields” is a software add-on for the NIFTy
“Numerical Information Field Theory” framework implementing Hamiltonian Monte
Carlo (HMC) sampling in Python. HMCF as well as NIFTy are designed to address
inference problems in high-dimensional spatially correlated setups such as image
They are optimized to deal with the typically high number of degrees of freedom.
HMCF adds an HMC sampler to NIFTy that automatically adjusts the many free
parameters steering the HMC sampling machinery. A wide variety of features
ensure efficient full-posterior sampling for high-dimensional inference
These features include integration step size adjustment, evaluation of the mass
matrix, convergence diagnostics, higher order symplectic integration and
simultaneous sampling of parameters and hyperparameters in Bayesian hierarchical
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment