An Efficient Implementation of Riemannian Manifold Hamiltonian Monte Carlo for Gaussian Process Models

Ulrich Paquet; Marco Fraccaro

An Efficient Implementation of Riemannian Manifold Hamiltonian Monte Carlo for Gaussian Process Models

Machine Learning 2018-10-30 v1 Machine Learning

Authors: Ulrich Paquet , Marco Fraccaro

Abstract

This technical report presents pseudo-code for a Riemannian manifold Hamiltonian Monte Carlo (RMHMC) method to efficiently simulate samples from $N$ -dimensional posterior distributions $p(x|y)$ , where $x \in R^N$ is drawn from a Gaussian Process (GP) prior, and observations $y_n$ are independent given $x_n$ . Sufficient technical and algorithmic details are provided for the implementation of RMHMC for distributions arising from GP priors.

Keywords

riemannian manifold hamiltonian monte carlo gaussian process posterior sampling markov chain monte carlo pseudo-code

Cite

@article{arxiv.1810.11893,
  title  = {An Efficient Implementation of Riemannian Manifold Hamiltonian Monte Carlo for Gaussian Process Models},
  author = {Ulrich Paquet and Marco Fraccaro},
  journal= {arXiv preprint arXiv:1810.11893},
  year   = {2018}
}

Comments

Technical report accompanying arXiv:1604.01972, "An Adaptive Resample-Move Algorithm for Estimating Normalizing Constants" (2016)

An Efficient Implementation of Riemannian Manifold Hamiltonian Monte Carlo for Gaussian Process Models

Abstract

Keywords

Cite

Comments

Related papers