CoChat
Sign Up
Cited by 23Galaxies: Formation, Evolution, PhenomenaAdvanced Neural Network ApplicationsAstronomical Observations and Instrumentation

Systematic biases when using deep neural networks for annotating large catalogs of astronomical images

Open access
Dhar, Sanchari, Shamir, Lior·2022·Astronomy and Computing
DOI 10.1016/j.ascom.2022.100545
Open this paper in CoChat — free account

Integrity signals

Registry

Crossref · 10.1016/j.ascom.2022.100545

Cites

2 of 2 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Deep learning based classifications are sensitive to the location of the galaxy in the sky.

    “For example, not properly understanding the information used can generate some biases, as demonstrated by Dhar & Shamir (2022) which show that deep learning based classifications are sensitive to the location of the galaxy in the sky.”

    From The Dawes Review 10: The impact of deep learning for the analysis of galaxy surveys (2023) · cites this paper as Dhar & Shamir (2022)

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Dhar, S., & Shamir, L.
Affiliations
Kansas State University (US)
Corresponding author
Shamir, Lior
ORCID
Shamir, Lior
Venue
Astronomy and Computing
Publisher
Elsevier BV
Published
2022-01
Volume
38
Pages
100545
ISSN
2213-1337 (ISSN-L)
DOI
10.1016/j.ascom.2022.100545
OpenAlex
W4206454747
Cited by
23
References
2
Citations by year
2026: 1 · 2025: 5 · 2024: 6 · 2023: 6 · 2022: 5
FWCI
2.20
Type
journal article
Language
en
Open access
Licence
Locations
  • Astronomy and Computing (acceptedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Funders
  • Maritime and Port Authority of Singapore
  • Yale University
  • Los Alamos National Laboratory
  • Space Telescope Science Institute
  • University of Washington
  • Max-Planck-Gesellschaft
  • American Institute of Physics
  • University of Tokyo
  • University of Virginia
  • University of Portsmouth
  • Ohio State University
  • Queen's University Belfast
  • Leibniz-Institut für Astrophysik Potsdam
  • University of Edinburgh
  • Eötvös Loránd Tudományegyetem
  • University of Hawai'i
  • Pennsylvania State University
  • Carnegie Mellon University
  • National Aeronautics and Space Administration (NNX08AR22G)
  • Durham University
  • University of Notre Dame
  • Alfred P. Sloan Foundation
  • University of Maryland
  • University of Wisconsin-Madison
  • Gordon and Betty Moore Foundation
  • Johns Hopkins University
  • Vanderbilt University
  • National Astronomical Observatories, Chinese Academy of Sciences
  • U.S. Department of Energy
  • National Science Foundation (AST-1903823, AST-1238877)
  • University of Utah
Topics
  • Physical Sciences › Physics and Astronomy › Astronomy and Astrophysics › Galaxies: Formation, Evolution, Phenomena
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Engineering › Computational Mechanics › Astronomical Observations and Instrumentation
Keywords
deep convolutional neural networks, systematic bias, galaxy morphology classification, elliptical galaxies, spiral galaxies, cosmic anisotropy, training set construction, digital sky surveys

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review