A Caltech Library Service

α-deep Probabilistic Inference (α-DPI): Efficient Uncertainty Quantification from Exoplanet Astrometry to Black Hole Feature Extraction

Sun, He and Bouman, Katherine L. and Tiede, Paul and Wang, Jason J. and Blunt, Sarah and Mawet, Dimitri (2022) α-deep Probabilistic Inference (α-DPI): Efficient Uncertainty Quantification from Exoplanet Astrometry to Black Hole Feature Extraction. Astrophysical Journal, 932 (2). Art. No. 99. ISSN 0004-637X. doi:10.3847/1538-4357/ac6be9.

[img] PDF - Published Version
Creative Commons Attribution.

[img] PDF - Submitted Version
Creative Commons Attribution.


Use this Persistent URL to link to this item:


Inference is crucial in modern astronomical research, where hidden astrophysical features and patterns are often estimated from indirect and noisy measurements. Inferring the posterior of hidden features, conditioned on the observed measurements, is essential for understanding the uncertainty of results and downstream scientific interpretations. Traditional approaches for posterior estimation include sampling-based methods and variational inference (VI). However, sampling-based methods are typically slow for high-dimensional inverse problems, while VI often lacks estimation accuracy. In this paper, we propose α-deep probabilistic inference, a deep learning framework that first learns an approximate posterior using α-divergence VI paired with a generative neural network, and then produces more accurate posterior samples through importance reweighting of the network samples. It inherits strengths from both sampling and VI methods: it is fast, accurate, and more scalable to high-dimensional problems than conventional sampling-based approaches. We apply our approach to two high-impact astronomical inference problems using real data: exoplanet astrometry and black hole feature extraction.

Item Type:Article
Related URLs:
URLURL TypeDescription Paper
Sun, He0000-0003-1526-6787
Bouman, Katherine L.0000-0003-0077-4367
Tiede, Paul0000-0003-3826-5648
Wang, Jason J.0000-0003-0774-6502
Blunt, Sarah0000-0002-3199-2888
Mawet, Dimitri0000-0002-8895-4735
Alternate Title:alpha-Deep Probabilistic Inference (alpha-DPI): efficient uncertainty quantification from exoplanet astrometry to black hole feature extraction
Additional Information:© 2022. The Author(s). Published by the American Astronomical Society. Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. Received 2022 January 20; revised 2022 April 29; accepted 2022 April 29; published 2022 June 20. This work was supported by NSF award 1935980, NSF award 2034306, NSF award 2048237, and Amazon AI4Science Fellowship. We thank the National Science Foundation (AST-1935980) for financial support of this work. This work has been supported in part by the Black Hole Initiative at Harvard University, which is funded by grants from the John Templeton Foundation and the Gordon and Betty Moore Foundation to Harvard University. This work was supported in part by Perimeter Institute for Theoretical Physics. Research at Perimeter Institute is supported by the Government of Canada through the Department of Innovation, Science and Economic Development Canada and by the Province of Ontario through the Ministry of Economic Development, Job Creation and Trade. J.W. and S.B. are supported by the Heising-Simons Foundation (including grant 2019-1698). The authors would also like to thank Shiro Ikeda for the helpful discussions. Facilities: Gemini Planet Imager (GPI) - , Event Horizon Telescope (EHT) - . Software: astropy (Astropy Collaboration et al. 2013, 2018), Cloudy (Ferland et al. 2013), Source Extractor (Bertin & Arnouts 1996) eht-imaging (Chael et al. 2018).
Group:Astronomy Department
Funding AgencyGrant Number
Amazon AI4Science FellowshipUNSPECIFIED
Black Hole InitiativeUNSPECIFIED
John Templeton FoundationUNSPECIFIED
Gordon and Betty Moore FoundationUNSPECIFIED
Perimeter Institute for Theoretical PhysicsUNSPECIFIED
Department of Innovation, Science and Economic Development (Canada)UNSPECIFIED
Ontario Ministry of Economic Development, Job Creation and TradeUNSPECIFIED
Heising-Simons Foundation2019-1698
Subject Keywords:Astrostatistics; Radio interferometry; Exoplanets; Black holes
Issue or Number:2
Classification Code:Unified Astronomy Thesaurus concepts: Astrostatistics (1882); Radio interferometry (1346); Exoplanets (498); Black holes (162)
Record Number:CaltechAUTHORS:20220125-215129959
Persistent URL:
Official Citation:He Sun et al 2022 ApJ 932 99
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:113103
Deposited By: George Porter
Deposited On:25 Jan 2022 22:14
Last Modified:15 Jul 2022 17:02

Repository Staff Only: item control page