A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow

Abstract We test the possibility of using a convolutional neural network to infer the inclination angle of a black hole directly from the incomplete image of the black hole’s shadow in the u v -plane. To this end, we develop a proof-of-concept network and use it to explicitly find how the error depends on the degree of coverage, type of input and coverage pattern. We arrive at a typical error of 10° at a level of absolute coverage 1% (for a pattern covering a central part of the u v -plane), 0.3% (pattern covering the central part and the periphery, the 0.3% referring to the central part only), and 14% (uniform pattern). These numbers refer to a network that takes both amplitude and phase of the visibility function as inputs. We find that this type of network works best in terms of the error itself and its distribution for different angles. In addition, the same type of network demonstrates similarly good performance on highly blurred images mimicking sources nearing being unresolved. In terms of coverage, the magnitude of the error does not change much as one goes from the central pattern to the uniform one. We argue that this may be due to the presence of a typical scale which can be mostly learned by the network from the central part alone.

[1]  D. Psaltis Testing general relativity with the Event Horizon Telescope , 2018, General Relativity and Gravitation.

[2]  Philippe Berger,et al.  A volumetric deep Convolutional Neural Network for simulation of dark matter halo catalogues , 2018, Monthly Notices of the Royal Astronomical Society.

[3]  M. Shea,et al.  Imaging black holes and jets with a VLBI array including multiple space-based telescopes , 2019, 1903.09539.

[4]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[5]  H. Falcke,et al.  The Size, Shape, and Scattering of Sagittarius A* at 86 GHz: First VLBI with ALMA , 2019, The Astrophysical Journal.

[6]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[7]  S. Shapiro,et al.  An Analysis of Variance Test for Normality (Complete Samples) , 1965 .

[8]  Daniel C. M. Palumbo,et al.  The Event Horizon General Relativistic Magnetohydrodynamic Code Comparison Project , 2019, The Astrophysical Journal Supplement Series.

[9]  Bruce A. Bassett,et al.  Painting galaxies into dark matter haloes using machine learning , 2017, 1712.03255.

[10]  M. Giersz,et al.  Finding black holes with black boxes – using machine learning to identify globular clusters with black hole subsystems , 2018, Monthly Notices of the Royal Astronomical Society.

[11]  Gaël Varoquaux,et al.  The NumPy Array: A Structure for Efficient Numerical Computation , 2011, Computing in Science & Engineering.

[12]  A. Niell,et al.  Event-horizon-scale structure in the supermassive black hole candidate at the Galactic Centre , 2008, Nature.

[13]  Reinhard Genzel,et al.  The galactic center massive black hole and nuclear star cluster , 2010, 1006.0064.

[14]  Tao An,et al.  The Chinese space Millimeter-wavelength VLBI array-A step toward imaging the most compact astronomical objects , 2014, 1403.5188.

[15]  J. Kneib,et al.  Deep convolutional neural networks as strong gravitational lens detectors , 2017, 1705.07132.

[16]  D. Clery Astronomy. Worldwide telescope aims to look into Milky Way galaxy's black heart. , 2012, Science.

[17]  John D. Hunter,et al.  Matplotlib: A 2D Graphics Environment , 2007, Computing in Science & Engineering.

[18]  Jean-Pierre Luminet,et al.  Image of a spherical black hole with thin accretion disk. , 1979 .

[19]  H. Falcke,et al.  Asymmetric structure in Sgr A* at 3 mm from closure phase measurements with VLBA, GBT and LMT , 2016, 1608.06515.

[20]  A. Michelson,et al.  Measurement of the Diameter of Alpha-Orionis by the Interferometer. , 1921, Proceedings of the National Academy of Sciences of the United States of America.

[21]  Max Welling,et al.  Data-driven Reconstruction of Gravitationally Lensed Galaxies Using Recurrent Inference Machines , 2019, The Astrophysical Journal.

[22]  Ayman Bin Kamruddin,et al.  A geometric crescent model for black hole images , 2013, 1306.3226.

[23]  Brian E. Granger,et al.  IPython: A System for Interactive Scientific Computing , 2007, Computing in Science & Engineering.

[24]  S. T. Timmer,et al.  First M87 Event Horizon Telescope Results. I. The Shadow of the Supermassive Black Hole , 2019, 1906.11238.

[25]  M. Wright,et al.  1.3 mm WAVELENGTH VLBI OF SAGITTARIUS A*: DETECTION OF TIME-VARIABLE EMISSION ON EVENT HORIZON SCALES , 2010, 1011.2472.

[26]  Daniel C. M. Palumbo,et al.  First M87 Event Horizon Telescope Results. VI. The Shadow and Mass of the Central Black Hole , 2019, The Astrophysical Journal.

[27]  Denis Donnelly,et al.  The fast Fourier transform for experimentalists. Part I. Concepts , 2005, Comput. Sci. Eng..

[28]  Daniel C. M. Palumbo,et al.  First M87 Event Horizon Telescope Results. III. Data Processing and Calibration , 2019, The Astrophysical Journal.

[29]  Zhi-qiang Shen,et al.  Space very long baseline interferometry in China , 2019 .

[30]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[31]  Laurence Perreault Levasseur,et al.  Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing , 2017, 1708.08843.

[32]  Jeremy D. Murphy,et al.  DYNAMICAL MEASUREMENTS OF BLACK HOLE MASSES IN FOUR BRIGHTEST CLUSTER GALAXIES AT 100 Mpc , 2012, 1203.1620.

[33]  John Salvatier,et al.  Theano: A Python framework for fast computation of mathematical expressions , 2016, ArXiv.

[34]  N. R. Napolitano,et al.  Finding strong gravitational lenses in the Kilo Degree Survey with Convolutional Neural Networks , 2017, 1702.07675.

[35]  Jean-Pierre Luminet,et al.  An Illustrated History of Black Hole Imaging : Personal Recollections (1972-2002) , 2019, 1902.11196.

[36]  Max Welling,et al.  Recurrent Inference Machines for Solving Inverse Problems , 2017, ArXiv.

[37]  Karl Glazebrook,et al.  Finding strong lenses in CFHTLS using convolutional neural networks , 2017, 1704.02744.

[38]  Christopher M. Bishop,et al.  Pattern Recognition and Machine Learning (Information Science and Statistics) , 2006 .

[39]  Asantha Cooray,et al.  LensFlow: A Convolutional Neural Network in Search of Strong Gravitational Lenses , 2017, 1705.05857.

[40]  X. Hong,et al.  Progress and Perspective of Space VLBI in China , 2019, 1901.07796.

[41]  B. Póczos,et al.  CMU DeepLens: deep learning for automatic image-based galaxy–galaxy strong lens finding , 2017, Monthly Notices of the Royal Astronomical Society.

[42]  J. Högbom,et al.  APERTURE SYNTHESIS WITH A NON-REGULAR DISTRIBUTION OF INTERFEROMETER BASELINES. Commentary , 1974 .

[43]  B. G. Clark An efficient implementation of the algorithm 'CLEAN' , 1980 .

[44]  L. Ho,et al.  Coevolution (Or Not) of Supermassive Black Holes and Host Galaxies: Supplemental Material , 2013, 1304.7762.

[45]  William H. Press,et al.  Numerical recipes in C. The art of scientific computing , 1987 .

[46]  Y. Alibert,et al.  Using deep neural networks to compute the mass of forming planets , 2019, Astronomy & Astrophysics.

[47]  Kevin A. Dudevoir,et al.  First M87 Event Horizon Telescope Results. II. Array and Instrumentation , 2019, 1906.11239.

[48]  Alan E. E. Rogers,et al.  Jet-Launching Structure Resolved Near the Supermassive Black Hole in M87 , 2012, Science.

[49]  François Chollet,et al.  Deep Learning with Python , 2017 .

[50]  Patrick Putzky,et al.  Analyzing interferometric observations of strong gravitational lenses with recurrent and convolutional neural networks , 2018, 1808.00011.

[51]  G. Swenson,et al.  Interferometry and Synthesis in Radio Astronomy , 1986 .

[52]  Katherine L. Bouman,et al.  Metrics and Motivations for Earth–Space VLBI: Time-resolving Sgr A* with the Event Horizon Telescope , 2019, The Astrophysical Journal.

[53]  Chih-Wei L. Huang,et al.  First M87 Event Horizon Telescope Results. IV. Imaging the Central Supermassive Black Hole , 2019 .

[54]  Daniel C. M. Palumbo,et al.  First M87 Event Horizon Telescope Results. V. Physical Origin of the Asymmetric Ring , 2019, The Astrophysical Journal.

[55]  R. Narayan,et al.  Maximum Entropy Image Restoration in Astronomy , 1986 .

[56]  S. F. Likhachev,et al.  Review of scientific topics for the Millimetron space observatory , 2014 .

[57]  Laurence Perreault Levasseur,et al.  Fast automated analysis of strong gravitational lenses with convolutional neural networks , 2017, Nature.

[58]  Jan-Jakob Sonke,et al.  Recurrent inference machines for reconstructing heterogeneous MRI data , 2019, Medical Image Anal..

[59]  Daniel C. M. Palumbo,et al.  THEMIS: A Parameter Estimation Framework for the Event Horizon Telescope , 2020, The Astrophysical Journal.

[60]  Alan E. E. Rogers,et al.  230 GHz VLBI OBSERVATIONS OF M87: EVENT‐HORIZON‐SCALE STRUCTURE DURING AN ENHANCED VERY‐HIGH‐ENERGY γ ?> ‐RAY STATE IN 2012 , 2015, 1505.03545.

[61]  Alan E. E. Rogers,et al.  Detection of Intrinsic Source Structure at ∼3 Schwarzschild Radii with Millimeter-VLBI Observations of SAGITTARIUS A* , 2018, 1805.09223.