Marginal unbiased score expansion and application to CMB lensing
Here, we present the marginal unbiased score expansion (MUSE) method, an algorithm for generic high-dimensional hierarchical Bayesian inference. MUSE performs approximate marginalization over arbitrary non-Gaussian latent parameter spaces, yielding Gaussianized asymptotically unbiased and near-optim...
Published in: | Physical Review D |
---|---|
Main Authors: | , |
Language: | unknown |
Published: |
2023
|
Subjects: | |
Online Access: | http://www.osti.gov/servlets/purl/1878325 https://www.osti.gov/biblio/1878325 https://doi.org/10.1103/physrevd.105.103531 |
id |
ftosti:oai:osti.gov:1878325 |
---|---|
record_format |
openpolar |
spelling |
ftosti:oai:osti.gov:1878325 2023-07-30T04:06:55+02:00 Marginal unbiased score expansion and application to CMB lensing Millea, Marius Seljak, Uroš 2023-05-25 application/pdf http://www.osti.gov/servlets/purl/1878325 https://www.osti.gov/biblio/1878325 https://doi.org/10.1103/physrevd.105.103531 unknown http://www.osti.gov/servlets/purl/1878325 https://www.osti.gov/biblio/1878325 https://doi.org/10.1103/physrevd.105.103531 doi:10.1103/physrevd.105.103531 72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS 2023 ftosti https://doi.org/10.1103/physrevd.105.103531 2023-07-11T10:13:45Z Here, we present the marginal unbiased score expansion (MUSE) method, an algorithm for generic high-dimensional hierarchical Bayesian inference. MUSE performs approximate marginalization over arbitrary non-Gaussian latent parameter spaces, yielding Gaussianized asymptotically unbiased and near-optimal constraints on global parameters of interest. It is computationally much cheaper than exact alternatives like Hamiltonian Monte Carlo (HMC), excelling on funnel problems which challenge HMC, and does not require any problem-specific user supervision like other approximate methods such as variational inference or many simulation-based inference methods. MUSE makes possible the first joint Bayesian estimation of the delensed Cosmic Microwave Background (CMB) power spectrum and gravitational lensing potential power spectrum, demonstrated here on a simulated data set as large as the upcoming South Pole Telescope 3G 1500 deg 2 survey, corresponding to a latent dimensionality of ~6 million and of order 100 global bandpower parameters. On a subset of the problem where an exact but more expensive HMC solution is feasible, we verify that MUSE yields nearly optimal results. We also demonstrate that existing spectrum-based forecasting tools which ignore pixel-masking underestimate predicted error bars by only ~10%. This method is a promising path forward for fast lensing and delensing analyses which will be necessary for future CMB experiments such as SPT-3G, Simons Observatory, or CMB-S4, and can complement or supersede existing HMC approaches. The success of MUSE on this challenging problem strengthens its case as a generic procedure for a broad class of high-dimensional inference problems. Other/Unknown Material South pole SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) South Pole Physical Review D 105 10 |
institution |
Open Polar |
collection |
SciTec Connect (Office of Scientific and Technical Information - OSTI, U.S. Department of Energy) |
op_collection_id |
ftosti |
language |
unknown |
topic |
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS |
spellingShingle |
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS Millea, Marius Seljak, Uroš Marginal unbiased score expansion and application to CMB lensing |
topic_facet |
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS |
description |
Here, we present the marginal unbiased score expansion (MUSE) method, an algorithm for generic high-dimensional hierarchical Bayesian inference. MUSE performs approximate marginalization over arbitrary non-Gaussian latent parameter spaces, yielding Gaussianized asymptotically unbiased and near-optimal constraints on global parameters of interest. It is computationally much cheaper than exact alternatives like Hamiltonian Monte Carlo (HMC), excelling on funnel problems which challenge HMC, and does not require any problem-specific user supervision like other approximate methods such as variational inference or many simulation-based inference methods. MUSE makes possible the first joint Bayesian estimation of the delensed Cosmic Microwave Background (CMB) power spectrum and gravitational lensing potential power spectrum, demonstrated here on a simulated data set as large as the upcoming South Pole Telescope 3G 1500 deg 2 survey, corresponding to a latent dimensionality of ~6 million and of order 100 global bandpower parameters. On a subset of the problem where an exact but more expensive HMC solution is feasible, we verify that MUSE yields nearly optimal results. We also demonstrate that existing spectrum-based forecasting tools which ignore pixel-masking underestimate predicted error bars by only ~10%. This method is a promising path forward for fast lensing and delensing analyses which will be necessary for future CMB experiments such as SPT-3G, Simons Observatory, or CMB-S4, and can complement or supersede existing HMC approaches. The success of MUSE on this challenging problem strengthens its case as a generic procedure for a broad class of high-dimensional inference problems. |
author |
Millea, Marius Seljak, Uroš |
author_facet |
Millea, Marius Seljak, Uroš |
author_sort |
Millea, Marius |
title |
Marginal unbiased score expansion and application to CMB lensing |
title_short |
Marginal unbiased score expansion and application to CMB lensing |
title_full |
Marginal unbiased score expansion and application to CMB lensing |
title_fullStr |
Marginal unbiased score expansion and application to CMB lensing |
title_full_unstemmed |
Marginal unbiased score expansion and application to CMB lensing |
title_sort |
marginal unbiased score expansion and application to cmb lensing |
publishDate |
2023 |
url |
http://www.osti.gov/servlets/purl/1878325 https://www.osti.gov/biblio/1878325 https://doi.org/10.1103/physrevd.105.103531 |
geographic |
South Pole |
geographic_facet |
South Pole |
genre |
South pole |
genre_facet |
South pole |
op_relation |
http://www.osti.gov/servlets/purl/1878325 https://www.osti.gov/biblio/1878325 https://doi.org/10.1103/physrevd.105.103531 doi:10.1103/physrevd.105.103531 |
op_doi |
https://doi.org/10.1103/physrevd.105.103531 |
container_title |
Physical Review D |
container_volume |
105 |
container_issue |
10 |
_version_ |
1772819875857694720 |