Deep Horizon; a machine learning network that recovers accreting black hole parameters

Abstract

The Event Horizon Telescope recently observed the first shadow of a black hole. Images like this can potentially be used to test or constrain theories of gravity and deepen the understanding in plasma physics at event horizon scales, which requires accurate parameter estimations. In this work, we present Deep Horizon, two convolutional deep neural networks that recover the physical parameters from images of black hole shadows. We investigate the effects of a limited telescope resolution and observations at higher frequencies. We trained two convolutional deep neural networks on a large image library of simulated mock data. The first network is a Bayesian deep neural regression network and is used to recover the viewing angle $i$, and position angle, mass accretion rate $\dot{M}$, electron heating prescription $R_{\mathrm{high}}$ and the black hole mass $M_{\mathrm{BH}}$. The second network is a classification network that recovers the black hole spin $a$. We find that with the current resolution of the Event Horizon Telescope, it is only possible to accurately recover a limited number of parameters of a static image, namely the mass and mass accretion rate. Since potential future space-based observing missions will operate at frequencies above 230 GHz, we also investigated the applicability of our network at a frequency of 690 GHz. The expected resolution of space-based missions is higher than the current resolution of the Event Horizon Telescope, and we show that Deep Horizon can accurately recover the parameters of simulated observations with a comparable resolution to such missions.

13 pages, 10 figures, 2 tables

References in corpus (19)

• The NumPy array: a structure for efficient numerical computation
• First M87 Event Horizon Telescope Results. I. The Shadow of the Supermassive Black Hole
• First M87 Event Horizon Telescope Results. VI. The Shadow and Mass of the Central Black Hole
• First M87 Event Horizon Telescope Results. IV. Imaging the Central Supermassive Black Hole
• First M87 Event Horizon Telescope Results. V. Physical Origin of the Asymmetric Ring
• First M87 Event Horizon Telescope Results. II. Array and Instrumentation
• First M87 Event Horizon Telescope Results. III. Data Processing and Calibration
• Black Hole Shadows, Photon Rings, and Lensing Rings
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• The Shadow of a Spherically Accreting Black Hole
• Star-galaxy Classification Using Deep Convolutional Neural Networks
• Uncertainties in Parameters Estimated with Neural Networks: Application to Strong Gravitational Lensing
• Electron and proton heating in trans-relativistic magnetic reconnection
• 230 GHz VLBI observations of M87: event-horizon-scale structure at the enhanced very-high-energy $\gamma$-ray state in 2012
• Robust Machine Learning Applied to Astronomical Datasets I: Star-Galaxy Classification of the SDSS DR3 Using Decision Trees
• Fundamental Parameters of Main-Sequence Stars in an Instant with Machine Learning
• Simulations of imaging the event horizon of Sagittarius A* from space
• Deep Learning Classification in Asteroseismology
• Effect of training characteristics on object classification: an application using Boosted Decision Trees

Cited by in corpus (17)

• TeraHertz Exploration and Zooming-in for Astrophysics (THEZA): ESA Voyage 2050 White Paper
• Silhouettes of invisible black holes
• Using space-VLBI to probe gravity around Sgr A*
• Black hole parameter estimation with synthetic Very Long Baseline Interferometry data from the ground and from space
• Using Machine Learning to Link Black Hole Accretion Flows with Spatially Resolved Polarimetric Observables
• An Event Horizon Imager (EHI) Mission Concept Utilizing Medium Earth Orbit Sub-mm Interferometry
• Black Hole Weather Forecasting with Deep Learning: A Pilot Study
• Imaging the event horizon of M87* from space on different timescales
• Deep learning inference with the Event Horizon Telescope II. The Zingularity framework for Bayesian artificial neural networks
• A proof-of-concept neural network for inferring parameters of a black hole from partial interferometric images of its shadow
• Observational properties of black hole in quantum fluctuation modified gravity
• Testing the Existence of Event Horizons against Rotating Reflecting Surfaces
• Characterization of black hole accretion through image moment invariants
• The science case and challenges of space-borne sub-millimeter interferometry
• Visible shapes of black holes M87* and SgrA*
• On Lightlike Geometry of Indefinite Sasakian Statistical Manifolds
• Parameter Inference of Black Hole Images using Deep Learning in Visibility Space