Accelerated Bayesian inference using deep learning. (arXiv:1903.10860v1 [astro-ph.CO])
<a href="http://arxiv.org/find/astro-ph/1/au:+Moss_A/0/1/0/all/0/1">Adam Moss</a>
We present a novel Bayesian inference tool that uses a neural network to
parameterise efficient Markov Chain Monte-Carlo (MCMC) proposals. The target
distribution is first transformed into a diagonal, unit variance Gaussian by a
series of non-linear, invertible, and non-volume preserving flows. Neural
networks are extremely expressive, and can transform complex targets to a
simple latent representation. Efficient proposals can then be made in this
space, and we demonstrate a high degree of mixing on several challenging
distributions. Parameter space can naturally be split into a block diagonal
speed hierarchy, allowing for fast exploration of subspaces where it is
inexpensive to evaluate the likelihood. Using this method, we develop a nested
MCMC sampler to perform Bayesian inference and model comparison, finding
excellent performance on highly curved and multi-modal analytic likelihoods. We
also test it on {em Planck} 2015 data, showing accurate parameter constraints,
and calculate the evidence for simple one-parameter extensions to LCDM in
$sim20$ dimensional parameter space. Our method has wide applicability to a
range of problems in astronomy and cosmology.
We present a novel Bayesian inference tool that uses a neural network to
parameterise efficient Markov Chain Monte-Carlo (MCMC) proposals. The target
distribution is first transformed into a diagonal, unit variance Gaussian by a
series of non-linear, invertible, and non-volume preserving flows. Neural
networks are extremely expressive, and can transform complex targets to a
simple latent representation. Efficient proposals can then be made in this
space, and we demonstrate a high degree of mixing on several challenging
distributions. Parameter space can naturally be split into a block diagonal
speed hierarchy, allowing for fast exploration of subspaces where it is
inexpensive to evaluate the likelihood. Using this method, we develop a nested
MCMC sampler to perform Bayesian inference and model comparison, finding
excellent performance on highly curved and multi-modal analytic likelihoods. We
also test it on {em Planck} 2015 data, showing accurate parameter constraints,
and calculate the evidence for simple one-parameter extensions to LCDM in
$sim20$ dimensional parameter space. Our method has wide applicability to a
range of problems in astronomy and cosmology.
http://arxiv.org/icons/sfx.gif
