TY - JOUR
T1 - Reconstructing probability distributions with Gaussian processes
AU - McClintock, Thomas
AU - Rozo, Eduardo
N1 - Funding Information:
TM thanks V. Miranda, Niall MacCrann, and Anzˇe Slosar for helpful discussion. ER was supported by the DOE grant DE-SC0015975, grant FG-2016-6443, and the Cottrell Scholar program of the Research Corporation for Science Advancement. ER would like to thank Ann Lee for useful conversations that helped spark interest in this work.
Publisher Copyright:
© 2019 Oxford University Press. All rights reserved.
PY - 2019/11/1
Y1 - 2019/11/1
N2 - Modern cosmological analyses constrain physical parameters using Markov Chain Monte Carlo (MCMC) or similar sampling techniques. Oftentimes, these techniques are computationally expensive to run and require up to thousands of CPU hours to complete. Here we present amethod for reconstructing the log-probability distributions of completed experiments from an existing chain (or any set of posterior samples). The reconstruction is performed using Gaussian process regression for interpolating the log-probability. This allows for easy resampling, importance sampling, marginalization, testing different samplers, investigating chain convergence, and other operations. As an example use case, we reconstruct the posterior distribution of the most recent Planck 2018 analysis. We then resample the posterior, and generate a newchainwith 40 times as many points in only 30min.Our likelihood reconstruction tool is made publicly available online.
AB - Modern cosmological analyses constrain physical parameters using Markov Chain Monte Carlo (MCMC) or similar sampling techniques. Oftentimes, these techniques are computationally expensive to run and require up to thousands of CPU hours to complete. Here we present amethod for reconstructing the log-probability distributions of completed experiments from an existing chain (or any set of posterior samples). The reconstruction is performed using Gaussian process regression for interpolating the log-probability. This allows for easy resampling, importance sampling, marginalization, testing different samplers, investigating chain convergence, and other operations. As an example use case, we reconstruct the posterior distribution of the most recent Planck 2018 analysis. We then resample the posterior, and generate a newchainwith 40 times as many points in only 30min.Our likelihood reconstruction tool is made publicly available online.
KW - Methods: Data analysis
UR - http://www.scopus.com/inward/record.url?scp=85075120931&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075120931&partnerID=8YFLogxK
U2 - 10.1093/mnras/stz2426
DO - 10.1093/mnras/stz2426
M3 - Article
AN - SCOPUS:85075120931
VL - 489
SP - 4155
EP - 4160
JO - Monthly Notices of the Royal Astronomical Society
JF - Monthly Notices of the Royal Astronomical Society
SN - 0035-8711
IS - 3
ER -