pith. sign in

arxiv: 2607.02110 · v1 · pith:XI2YWWF4new · submitted 2026-07-02 · ⚛️ physics.hist-ph · astro-ph.HE· astro-ph.IM· gr-qc

Black Boxes in Black Hole Imaging

Juliusz Doboszewski , Jamee Elder This is my paper

Pith reviewed 2026-07-03 01:40 UTC · model grok-4.3

classification ⚛️ physics.hist-ph astro-ph.HEastro-ph.IMgr-qc
keywords epistemic opacityblack hole imagingGRMHD modelsSagittarius A*machine learningEvent Horizon Telescopeinferential frameworkscomputer simulations
0
0 comments X

The pith

GRMHD models of Sagittarius A* exhibit a problematic epistemic opacity that signals limits in current understanding and restricts machine learning applications.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines epistemic opacity arising from computer simulations and machine learning in black hole imaging. It argues that certain forms of opacity, including those from machine learning, need not undermine the reliability of inferences when those methods sit inside a broader inferential framework. The authors propose conditions under which such opacity can remain compatible with reliable results for the Event Horizon Telescope. They identify one case, however, where opacity is problematic: GRMHD models of Sagittarius A* cannot be treated this way. This opacity indicates gaps in present understanding of the source models and thereby limits how machine learning can be deployed in future observations.

Core claim

While opacity from simulations or machine learning can be compatible with reliable inference when embedded in a suitable broader inferential framework, GRMHD models of Sagittarius A* possess a distinct problematic form of opacity. This form signals limitations of current understanding of the models of this source and constrains the potential uses of ML models in future observations.

What carries the argument

Epistemic opacity distinguished into forms that can be accommodated within broader inferential frameworks versus a problematic form present in GRMHD models of Sagittarius A*.

If this is right

  • Opaque methods can remain useful for next-generation Event Horizon Telescope observations when the proposed conditions for compatibility with reliable inference hold.
  • The problematic opacity in GRMHD models of Sagittarius A* directly constrains how machine learning methods can be applied to future observations of this source.
  • The identified opacity indicates specific gaps in present understanding of the models used for Sagittarius A*.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the distinction between benign and problematic opacity holds, similar analysis could be applied to GRMHD modeling of other black hole sources.
  • Clarifying the exact conditions for non-problematic opacity might allow targeted improvements to simulation pipelines rather than blanket avoidance of opaque components.

Load-bearing premise

The assumption that the opacity in GRMHD models of Sagittarius A* is a distinct problematic kind signaling fundamental model limitations, rather than a form that could be mitigated within a broader inferential framework.

What would settle it

A demonstration that the opacity in GRMHD models of Sagittarius A* can be rendered non-problematic by embedding those models in a broader inferential framework that satisfies the paper's proposed conditions would falsify the central claim.

read the original abstract

We investigate the epistemic opacity of computer simulations and machine learning methods in the context of black hole imaging. We argue that there are forms of opacity-including opacity resulting from the use of machine learning-which do not need to affect the reliability of an inference when it is seen as a part of a broader inferential framework. We propose conditions under which that can plausibly be the case, and discuss how opaque methods can be useful in the context of the (next generation) Event Horizon Telescope. However, we also argue that at least one problematic form of opacity is currently present in black hole imaging: GRMHD models of Sagittarius A* are opaque. This form of opacity signals the limitations of current understanding of the models of this source, and constrains the potential uses of ML models in future observations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper claims that epistemic opacity in simulations and ML methods for black hole imaging (e.g., EHT) need not undermine reliability when embedded in a broader inferential framework, and proposes conditions under which this holds. It applies this to argue that GRMHD models of Sagittarius A* exhibit a distinct, problematic form of opacity signaling fundamental limitations in current understanding, which in turn constrains future ML applications.

Significance. If the proposed conditions for acceptable opacity are made operational and the GRMHD case is shown to violate them via concrete model components, the work would offer a philosophically grounded yet practically relevant framework for evaluating reliability in astrophysical modeling, particularly for next-generation EHT observations. The paper's engagement with real scientific practice (Sgr A* imaging) is a strength.

major comments (2)
  1. [Abstract and discussion of GRMHD models] The distinction between mitigable and problematic opacity is asserted in the abstract and developed conceptually, but no explicit criteria or mapping is provided from specific GRMHD elements (turbulence closures, radiative transfer assumptions, or parameter choices) to violations of the proposed conditions. This leaves the central claim that GRMHD opacity 'signals the limitations of current understanding' unsupported by the general framework.
  2. [Sections proposing conditions for acceptable opacity] The conditions under which opacity is acceptable when part of a broader inferential framework are proposed but not tested against the GRMHD Sgr A* case with sufficient rigor to establish why this instance cannot be accommodated (unlike other simulation/ML opacities discussed). Without this, the constraint on ML uses does not follow.
minor comments (1)
  1. [Introduction] Clarify notation for 'opacity' early on to distinguish epistemic from physical senses, as the two are used in the black hole imaging context.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their detailed and constructive report. The comments highlight important areas where the connection between our proposed framework and its application to GRMHD models of Sgr A* can be strengthened. We respond to each major comment below and indicate the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract and discussion of GRMHD models] The distinction between mitigable and problematic opacity is asserted in the abstract and developed conceptually, but no explicit criteria or mapping is provided from specific GRMHD elements (turbulence closures, radiative transfer assumptions, or parameter choices) to violations of the proposed conditions. This leaves the central claim that GRMHD opacity 'signals the limitations of current understanding' unsupported by the general framework.

    Authors: The manuscript develops the conditions conceptually in the sections on inferential frameworks, and applies them to GRMHD by noting that the opacity arises from elements like turbulence closures that lack independent empirical validation for Sgr A*. However, we accept that an explicit mapping would make the argument more robust. In the revised version, we will add a dedicated paragraph or table that maps each condition to specific GRMHD components, showing how they fail to satisfy the requirements for mitigable opacity. revision: yes

  2. Referee: [Sections proposing conditions for acceptable opacity] The conditions under which opacity is acceptable when part of a broader inferential framework are proposed but not tested against the GRMHD Sgr A* case with sufficient rigor to establish why this instance cannot be accommodated (unlike other simulation/ML opacities discussed). Without this, the constraint on ML uses does not follow.

    Authors: We test the conditions by arguing that for GRMHD, the opacity cannot be accommodated within the current inferential framework because it reflects gaps in the underlying physical models rather than mere computational intractability. This is distinct from the ML cases where the broader EHT pipeline provides constraints. To address the concern about rigor, we will expand the relevant sections with a more systematic evaluation, including references to specific studies on GRMHD limitations for Sgr A*. revision: partial

Circularity Check

0 steps flagged

No circularity; conceptual distinctions are self-contained

full rationale

The paper advances a philosophical argument distinguishing opacity types that preserve inferential reliability within broader frameworks from a problematic instance in GRMHD models of Sgr A*. No equations, fitted parameters, or self-referential definitions appear in the provided abstract or described structure. Claims rest on proposed conditions and EHT discussion rather than any reduction of outputs to inputs by construction or via load-bearing self-citation chains. The derivation is therefore independent of the patterns that would indicate circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The paper rests on domain assumptions about what constitutes epistemic opacity and reliability in scientific inference, drawn from philosophy of science; no free parameters or invented entities are evident from the abstract.

axioms (2)
  • domain assumption Opacity from machine learning or simulations can be compatible with reliable inference when embedded in a broader inferential framework under certain conditions.
    This premise underpins the claim that some opacity does not affect reliability; invoked in the discussion of conditions for acceptable opacity.
  • domain assumption GRMHD models of Sgr A* exhibit a form of opacity that signals limitations in current physical understanding.
    Central to the negative claim about current models; stated directly in the abstract.

pith-pipeline@v0.9.1-grok · 5663 in / 1374 out tokens · 15328 ms · 2026-07-03T01:40:33.631663+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

262 extracted references · 48 canonical work pages · 1 internal anchor

  1. [1]

    Nature , number =

    Evidence of the pair-instability gap from black-hole masses , volume =. Nature , number =. 2026 , author =

    2026

  2. [2]

    How the Machine ‘Thinks’: Understanding Opacity in Machine Learning Algorithms , volume =

    Burrell, Jenna , address =. How the Machine ‘Thinks’: Understanding Opacity in Machine Learning Algorithms , volume =. Big Data & Society , number =

  3. [3]

    and Borchers, Angela and Fishbach, Maya and Ye, Claire S

    Passenger, Lachlan and Banagiri, Sharan and Thrane, Eric and Lasky, Paul D. and Borchers, Angela and Fishbach, Maya and Ye, Claire S. , title =. doi:10.3847/1538-4357/ae4358 , year =

    work page doi:10.3847/1538-4357/ae4358

  4. [4]

    Beyond generalization: a theory of robustness in machine learning , volume =

    Freiesleben, Timo and Grote, Thomas , address =. Beyond generalization: a theory of robustness in machine learning , volume =. Synthese , number =

  5. [5]

    Philosophy of Science , author=

    Inductive Risk, Understanding, and Opaque Machine Learning Models , volume=. Philosophy of Science , author=. 2022 , pages=. doi:10.1017/psa.2022.62 , number=

    work page doi:10.1017/psa.2022.62 2022

  6. [6]

    Canadian Journal of Philosophy , author=

    On the Opacity of Deep Neural Networks , volume=. Canadian Journal of Philosophy , author=. 2023 , pages=. doi:10.1017/can.2024.1 , number=

    work page doi:10.1017/can.2024.1 2023

  7. [7]

    Studies in History and Philosophy of Science Part A , volume =

    Variety of Evidence in Multimessenger Astronomy , year =. Studies in History and Philosophy of Science Part A , volume =. doi:10.1016/j.shpsa.2022.05.006 , author =

    work page doi:10.1016/j.shpsa.2022.05.006 2022

  8. [8]

    and Johnson, Michael D

    Bouman, Katherine L. and Johnson, Michael D. and Zoran, Daniel and Fish, Vincent L. and Doeleman, Sheperd S. and Freeman, William T. , booktitle=. Computational Imaging for VLBI Image Reconstruction , year=

  9. [9]

    Echoes from the Abyss: Tentative evidence for Planck-scale structure at black hole horizons , Volume =

    Abedi, Jahed and Dykaar, Hannah and Afshordi, Niayesh , Journal =. Echoes from the Abyss: Tentative evidence for Planck-scale structure at black hole horizons , Volume =

  10. [10]

    Nature Machine Intelligence , year =

    Rudin, Cynthia , title =. Nature Machine Intelligence , year =

  11. [11]

    , title =

    Boge, Florian J. , title =. Minds & Machines , year =

  12. [12]

    Prescod-Weinstein, Chanda , address =. The. The

  13. [13]

    and Akiyama, Kazunori and Blackburn, Lindy and Bouman, Katherine L

    Johnson, Michael D. and Akiyama, Kazunori and Blackburn, Lindy and Bouman, Katherine L. and Broderick, Avery E. and Cardoso, Vitor and Fender, Rob P. and Fromm, Christian M. and Galison, Peter and Gómez, José L. and Haggard, Daryl and Lister, Matthew L. and Lobanov, Andrei P. and Markoff, Sera and Narayan, Ramesh and Natarajan, Priyamvada and Nichols, Tif...

    2023

  14. [14]

    Historical Studies in the Natural Sciences , volume=

    Instruments of science or conquest? Neocolonialism and modern American astronomy , author=. Historical Studies in the Natural Sciences , volume=. 2017 , publisher=

    2017

  15. [15]

    arXiv preprint arXiv:2001.00970 , year=

    A Native Hawaiian-led summary of the current impact of constructing the Thirty Meter Telescope on Maunakea , author=. arXiv preprint arXiv:2001.00970 , year=

    work page arXiv 2001

  16. [16]

    and Martens, Niels C

    Marcoci, Alexandru and Thresher, Ann C. and Martens, Niels C. M. and Galison, Peter and Doeleman, Sheperd S. and Johnson, Michael D. , address =. Big. Nature human behaviour , number =

  17. [17]

    Philosophy of Science , author=

    Transparency in Complex Computational Systems , volume=. Philosophy of Science , author=. 2020 , pages=. doi:10.1086/709729 , number=

    work page doi:10.1086/709729 2020

  18. [18]

    2024 , url =

    Theory-mediated detection of novel phenomena in astrophysics: the case of the photon ring , author=. 2024 , url =

    2024

  19. [19]

    and Pesce, Dominic W

    Broderick, Avery E. and Pesce, Dominic W. and Gold, Roman and Tiede, Paul and Pu, Hung-Yi and Anantua, Richard and Britzen, Silke and Ceccobello, Chiara and Chatterjee, Koushik and Chen, Yongjun and Conroy, Nicholas S. and Crew, Geoffrey B. and Cruz-Osorio, Alejandro and Cui, Yuzhu and Doeleman, Sheperd S. and Emami, Razieh and Farah, Joseph and Fromm, Ch...

    2022

  20. [20]

    Philosophy of Science , volume=

    Model evaluation: An adequacy-for-purpose view , author=. Philosophy of Science , volume=. 2020 , publisher=

    2020

  21. [21]

    European Journal for Philosophy of Science , volume=

    Data models, representation and adequacy-for-purpose , author=. European Journal for Philosophy of Science , volume=. 2021 , publisher=

    2021

  22. [22]

    Synthese , number =

    Using models to correct data: paleodiversity and the fossil record , volume =. Synthese , number =. 2021 , author =

    2021

  23. [23]

    Synthese , volume=

    Conceptual challenges for interpretable machine learning , author=. Synthese , volume=. 2022 , publisher=

    2022

  24. [24]

    Philosophy Compass , volume=

    Philosophy of science at sea: Clarifying the interpretability of machine learning , author=. Philosophy Compass , volume=. 2022 , publisher=

    2022

  25. [25]

    The Astrophysical Journal , volume=

    -Deep Probabilistic Inference ( -DPI): efficient uncertainty quantification from exoplanet astrometry to black hole feature extraction , author=. The Astrophysical Journal , volume=. 2022 , publisher=

    2022

  26. [26]

    The Astrophysical Journal Supplement Series , volume=

    The event horizon general relativistic magnetohydrodynamic code comparison project , author=. The Astrophysical Journal Supplement Series , volume=. 2019 , publisher=

    2019

  27. [27]

    Monthly Notices of the Royal Astronomical Society , volume=

    Comparison of the ion-to-electron temperature ratio prescription: GRMHD simulations with electron thermodynamics , author=. Monthly Notices of the Royal Astronomical Society , volume=. 2021 , publisher=

    2021

  28. [28]

    arXiv preprint arXiv:2504.21840 , year=

    Parameter Inference of Black Hole Images using Deep Learning in Visibility Space , author=. arXiv preprint arXiv:2504.21840 , year=

    work page arXiv

  29. [29]

    IEEE Signal Processing Magazine , volume=

    Algorithm unrolling: Interpretable, efficient deep learning for signal and image processing , author=. IEEE Signal Processing Magazine , volume=. 2021 , publisher=

    2021

  30. [30]

    Technology in Society , volume=

    Sustaining local opposition to Big Science: A case study of the Thirty Meter Telescope controversy , author=. Technology in Society , volume=. 2024 , publisher=

    2024

  31. [31]

    Ethics, Policy & Environment , pages=

    A Fork on the Road to Knowledge at Mauna Kea: The Thirty Meter Telescope, Perspectivalism, and Epistemic Injustice , author=. Ethics, Policy & Environment , pages=. 2025 , publisher=

    2025

  32. [32]

    Philosophy of Science , year =

    Dethier, Corey , title =. Philosophy of Science , year =

  33. [33]

    arXiv preprint arXiv:2403.05452 , year=

    The R2D2 deep neural network series paradigm for fast precision imaging in radio astronomy , author=. arXiv preprint arXiv:2403.05452 , year=

    work page arXiv

  34. [34]

    Identifying asteroid trails in Hubble Space Telescope images , author=

    Hubble Asteroid Hunter-I. Identifying asteroid trails in Hubble Space Telescope images , author=. Astronomy & Astrophysics , volume=. 2022 , publisher=

    2022

  35. [35]

    The Astrophysical Journal , volume=

    Cosmology with one galaxy? , author=. The Astrophysical Journal , volume=. 2022 , publisher=

    2022

  36. [36]

    Monthly Notices of the Royal Astronomical Society , volume =

    Lockhart, Will and Gralla, Samuel E , title =. Monthly Notices of the Royal Astronomical Society , volume =. 2022 , month =

    2022

  37. [37]

    Galaxies , pages =

    Tiede, Paul and Johnson, Michael D and Pesce, Dominic W and Palumbo, Daniel C M and Chang, Dominic O and Galison, Peter , keywords =. Galaxies , pages =. 2022 , title =

    2022

  38. [38]

    Johnson and Alexandru Lupsasca and Andrew Strominger and George N

    Michael D. Johnson and Alexandru Lupsasca and Andrew Strominger and George N. Wong and Shahar Hadar and Daniel Kapec and Ramesh Narayan and Andrew Chael and Charles F. Gammie and Peter Galison and Daniel C. M. Palumbo and Sheperd S. Doeleman and Lindy Blackburn and Maciek Wielgus and Dominic W. Pesce and Joseph R. Farah and James M. Moran , title =. Scien...

    work page doi:10.1126/sciadv.aaz1310 2020

  39. [39]

    First M87 event horizon telescope results. I. The shadow of the supermassive black hole , author=. The Astrophysical Journal Letters , publisher =. 2019 , doi =

    2019

  40. [40]

    First M87 event horizon telescope results. II. Array and instrumentation , author=. The Astrophysical Journal Letters , volume=. 2019 , doi =

    2019

  41. [41]

    First M87 event horizon telescope results. III. Data processing and calibration , author=. The Astrophysical Journal Letters , volume=. 2019 , doi =

    2019

  42. [42]

    First M87 event horizon telescope results. IV. Imaging the central supermassive black hole , author=. The Astrophysical Journal Letters , volume=. 2019 , doi =

    2019

  43. [43]

    First M87 event horizon telescope results. V. Physical origin of the asymmetric ring , author=. The Astrophysical Journal Letters , volume=. 2019 , doi =

    2019

  44. [44]

    First M87 Event Horizon Telescope Results. VI. The Shadow and Mass of the Central Black Hole , journal =. doi:10.3847/2041-8213/ab1141 , year =

    work page doi:10.3847/2041-8213/ab1141 2041

  45. [45]

    First M87 Event Horizon Telescope Results. VII. Polarization of the Ring , author=. The Astrophysical Journal Letters , volume=. 2021 , doi =

    2021

  46. [46]

    First M87 Event Horizon Telescope Results. VIII. Magnetic Field Structure near The Event Horizon , author=. The Astrophysical Journal Letters , volume=. 2021 , doi =

    2021

  47. [47]

    Observations, calibration, imaging, and analysis , author=

    The persistent shadow of the supermassive black hole of M 87-I. Observations, calibration, imaging, and analysis , author=. Astronomy & Astrophysics , volume=. 2024 , publisher=

    2024

  48. [48]

    First Sagittarius A* Event Horizon Telescope Results. I. The Shadow of the Supermassive Black Hole in the Center of the Milky Way , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6674 ,

    work page doi:10.3847/2041-8213/ac6674 2022

  49. [49]

    First Sagittarius A* Event Horizon Telescope Results. II. EHT and Multiwavelength Observations, Data Processing, and Calibration , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6675 ,

    work page doi:10.3847/2041-8213/ac6675 2022

  50. [50]

    First Sagittarius A* Event Horizon Telescope Results. III. Imaging of the Galactic Center Supermassive Black Hole , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6429 ,

    work page doi:10.3847/2041-8213/ac6429 2022

  51. [51]

    First Sagittarius A* Event Horizon Telescope Results. IV. Variability, Morphology, and Black Hole Mass , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6736 ,

    work page doi:10.3847/2041-8213/ac6736 2022

  52. [52]

    First Sagittarius A* Event Horizon Telescope Results. V. Testing Astrophysical Models of the Galactic Center Black Hole , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6672 ,

    work page doi:10.3847/2041-8213/ac6672 2022

  53. [53]

    First Sagittarius A* Event Horizon Telescope Results. VI. Testing the Black Hole Metric , journal =. 2022 , publisher =. doi:10.3847/2041-8213/ac6756 ,

    work page doi:10.3847/2041-8213/ac6756 2022

  54. [54]

    First sagittarius a* event horizon telescope results. vii. polarization of the ring , author=. The Astrophysical Journal Letters , volume=. 2024 , publisher=

    2024

  55. [55]

    First sagittarius a* event horizon telescope results. viii. physical interpretation of the polarized ring , author=. The Astrophysical Journal Letters , volume=. 2024 , publisher=

    2024

  56. [56]

    and others , TITLE =

    Johnson, Michael D. and others , TITLE =. Galaxies , VOLUME =. 2023 , NUMBER =

    2023

  57. [57]

    Foundations of physics , number =

    Dark Matter Realism , volume =. Foundations of physics , number =. 2022 , author =

    2022

  58. [58]

    The British journal for the philosophy of science , number =

    Prediction and Explanation in Historical Natural Science , volume =. The British journal for the philosophy of science , number =. 2011 , author =

    2011

  59. [59]

    2013 , booktitle =

    Common cause explanation and the search for a smoking gun , author =. 2013 , booktitle =

    2013

  60. [60]

    A Tale of Two Dark Matter Concepts: The Relation Between Cosmology and High-energy Physics in the Context of Dark Matter Research , year =

    De Baerdemaeker, Siska , booktitle =. A Tale of Two Dark Matter Concepts: The Relation Between Cosmology and High-energy Physics in the Context of Dark Matter Research , year =

  61. [61]

    Galison, Peter and Doboszewski, Juliusz and Elder, Jamee and Martens, Niels C. M. and others , TITLE =. Galaxies , VOLUME =. 2023 , NUMBER =

    2023

  62. [62]

    The Astrophysical Journal Letters , volume=

    The Image of the M87 Black Hole Reconstructed with PRIMO , author=. The Astrophysical Journal Letters , volume=. 2023 , publisher=

    2023

  63. [63]

    The Astrophysical Journal , volume=

    A Red-noise Eigenbasis for the Reconstruction of Blobby Images , author=. The Astrophysical Journal , volume=. 2022 , publisher=

    2022

  64. [64]

    Philosophy of science , number =

    The Dark Galaxy Hypothesis , volume =. Philosophy of science , number =. 2018 , author =

    2018

  65. [65]

    Philosophy of Science , author=

    Contingency and History , volume=. Philosophy of Science , author=. 2016 , pages=. doi:10.1086/687260 , number=

    work page doi:10.1086/687260 2016

  66. [66]

    2016 , note =

    What are narratives good for? , journal =. 2016 , note =. doi:https://doi.org/10.1016/j.shpsc.2015.12.016 , author =

    work page doi:10.1016/j.shpsc.2015.12.016 2016

  67. [67]

    In defence of story-telling , volume =

    Currie, Adrian and Sterelny, Kim , address =. In defence of story-telling , volume =. Studies in history and philosophy of science. Part A , pages =

  68. [68]

    Scientific Knowledge and the Deep Past : History Matters , year =

    Currie, Adrian , address =. Scientific Knowledge and the Deep Past : History Matters , year =. Scientific Knowledge and the Deep Past : History Matters , publisher =

  69. [69]

    Studies in history and philosophy of science

    Testing times: regularities in the historical sciences , volume =. Studies in history and philosophy of science. Part C, Studies in history and philosophy of biological and biomedical sciences , number =. 2008 , author =

    2008

  70. [70]

    , address =

    Turner, Derek D. , address =. Making prehistory : historical science and the scientific realism debate , year =

  71. [71]

    Principal-component interferometric modeling (PRIMO), an algorithm for EHT data. I. Reconstructing images from simulated EHT observations , author=. The Astrophysical Journal , volume=. 2023 , publisher=

    2023

  72. [72]

    The Astrophysical Journal , volume=

    A General relativistic null hypothesis test with event horizon telescope observations of the black hole shadow in Sgr A , author=. The Astrophysical Journal , volume=. 2015 , publisher=

    2015

  73. [73]

    arXiv preprint arXiv:2408.10322 , year=

    Theoretical Foundation of Black Hole Image Reconstruction using PRIMO , author=. arXiv preprint arXiv:2408.10322 , year=

    work page arXiv

  74. [74]

    Astronomy & Astrophysics , volume=

    Deep-learning-based radiointerferometric imaging with GAN-aided training , author=. Astronomy & Astrophysics , volume=. 2023 , publisher=

    2023

  75. [75]

    Monthly Notices of the Royal Astronomical Society , volume=

    Deep radio-interferometric imaging with POLISH: DSA-2000 and weak lensing , author=. Monthly Notices of the Royal Astronomical Society , volume=. 2022 , publisher=

    2000

  76. [76]

    Astronomy & Astrophysics , volume=

    Deep learning-based imaging in radio interferometry , author=. Astronomy & Astrophysics , volume=. 2022 , publisher=

    2022

  77. [77]

    The Astrophysical Journal Letters , volume=

    First AI for deep super-resolution wide-field imaging in radio astronomy: unveiling structure in ESO 137-006 , author=. The Astrophysical Journal Letters , volume=. 2022 , publisher=

    2022

  78. [78]

    Erkenntnis , year =

    The Unity of Robustness: Why Agreement Across Model Reports is Just as Valuable as Agreement Among Experiments , author =. Erkenntnis , year =

  79. [79]

    Monthly Notices of the Royal Astronomical Society , volume =

    Jayasinghe, T and others , title = ". Monthly Notices of the Royal Astronomical Society , volume =. 2021 , month =

    2021

  80. [80]

    Advances in Space Research , volume=

    Space VLBI: from first ideas to operational missions , author=. Advances in Space Research , volume=. 2020 , publisher=

    2020

Showing first 80 references.