Characterizing the velocity of a wandering black hole and properties of the surrounding medium using convolutional neural networks

Abstract

We present a method for estimating the velocity of a wandering black hole and the equation of state for the gas around, based on a catalog of numerical simulations. The method uses machine learning methods based on convolutional neural networks applied to the classification of images resulting from numerical simulations. Specifically we focus on the supersonic velocity regime and choose the direction of the black hole to be parallel to its spin. We build a catalog of 900 simulations by numerically solving Euler's equations onto the fixed space-time background of a black hole, for two parameters: the adiabatic index $\Gamma$ with values in the range [1.1, 5/3], and the asymptotic relative velocity of the black hole with respect to the surroundings $v_{\infty }$, with values within $[0.2,0.8]c$. For each simulation we produce a 2D image of the gas density once the process of accretion has approached a stationary regime. The results obtained show that the implemented Convolutional Neural Networks are capable to classify correctly the adiabatic index $87.78\%$ of the time within an uncertainty of $\pm 0.0284$ while the prediction of the velocity is correct $96.67\%$ of the times within an uncertainty of $\pm 0.03c$. We expect that this combination of a massive number of numerical simulations and machine learning methods will help analyze more complicated scenarios related to future high resolution observations of black holes, like those from the Event Horizon Telescope.

5 RevTex pages. Published in Physical Review D

References in corpus (17)

• Event-horizon-scale structure in the supermassive black hole candidate at the Galactic Centre
• Large Merger Recoils and Spin Flips From Generic Black-Hole Binaries
• Total recoil: the maximum kick from nonspinning black-hole binary inspiral
• Jet Launching Structure Resolved Near the Supermassive Black Hole in M87
• Supermassive recoil velocities for binary black-hole mergers with antialigned spins
• Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation: Results with Advanced LIGO Data
• Massive Black Hole Binary Mergers in Dynamical Galactic Environments
• The old nuclear star cluster in the Milky Way: dynamics, mass, statistical parallax, and black hole mass
• Dynamical friction and the evolution of Supermassive Black hole Binaries: the final hundred-parsec problem
• The puzzling case of the radio-loud QSO 3C 186: a gravitational wave recoiling black hole in a young radio source?
• Is the flip-flop behaviour of accretion shock cones on to black holes an effect of coordinates?
• Axisymmetric Bondi-Hoyle accretion onto a Schwarzschild Black Hole: shock cone vibrations
• Kinematics of Ultra-High-Velocity Gas in the Expanding Molecular Shell adjacent to the W44 Supernova Remnant
• Modeling the Black hole Merger of QSO 3C 186
• Parameter estimates in binary black hole collisions using neural networks
• Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation with LIGO Data
• Classifying initial conditions of long GRBs modeled with relativistic radiation hydrodynamics

Cited by in corpus (12)

• Surveying the reach and maturity of machine learning and artificial intelligence in astronomy
• Gravitational Wave Denoising of Binary Black Hole Mergers with Deep Learning
• Fusing numerical relativity and deep learning to detect higher-order multipole waveforms from eccentric binary black hole mergers
• Dynamics of Intermediate-Mass Black Holes Wandering in the Milky Way Galaxy Using the Illustris TNG50 Simulation
• Detectability of Wandering Intermediate-Mass Black Holes in the Milky Way Galaxy from Radio to X-rays
• Orbital and Radiative Properties of Wandering Intermediate-Mass Black Holes in the ASTRID Simulation
• Linear perturbations of low angular momentum accretion flow in the Kerr metric and the corresponding emergent gravity phenomena
• Deep learning merger masses estimation from gravitational waves signals in the frequency domain
• Classification of a black hole spin out of its shadow using support vector machines
• Accretion of supersonic magnetized winds onto black holes
• Tracing black hole and galaxy growth across environments since cosmic noon
• Machine Learning for Nanohertz Gravitational Wave Detection and Parameter Estimation with Pulsar Timing Array