Abstract:In this paper, we show how the quasinormal modes (QNMs) arise from the perturbations of massive scalar fields propagating in the curved background by using the artificial neural networks. To this end, we architect a special algorithm for the feedforward neural network method (FNNM) to compute the QNMs complying with the certain types of boundary conditions. To test the reliability of the method, we consider two black hole spacetimes whose QNMs are well known: [Formula: see text] pure de Sitter (dS) and five-di… Show more
“…Mathematically, the artificial neural network is an optimization scheme that can be adopted to solve eigenvalue problems [87]. In [88], the method was adopted for black hole QNMs and exercised for four-dimensional pure dS and five-dimensional Schwarzschild AdS black holes. Good agreement was manifestly obtained when compared with other methods.…”
Section: Further Discussion and Concluding Remarksmentioning
Motivated by the substantial instability of the fundamental and high-overtone quasinormal modes, recent developments regarding the notion of black hole pseudospectrum call for numerical results with unprecedented precision.
This work generalizes and improves the matrix method for black hole quasinormal modes to higher orders, specifically aiming at a class of perturbations to the metric featured by discontinuity intimately associated with the quasinormal mode structural instability.
The approach is based on the mock-Chebyshev grid, which guarantees its convergence in the degree of the interpolant.
In practice, solving for black hole quasinormal modes is a formidable task.
The presence of discontinuity poses a further difficulty so that many well-known approaches cannot be employed straightforwardly.
Compared with other viable methods, the modified matrix method is competent in speed and accuracy. 
Therefore, the method serves as a helpful gadget for relevant studies.
“…Mathematically, the artificial neural network is an optimization scheme that can be adopted to solve eigenvalue problems [87]. In [88], the method was adopted for black hole QNMs and exercised for four-dimensional pure dS and five-dimensional Schwarzschild AdS black holes. Good agreement was manifestly obtained when compared with other methods.…”
Section: Further Discussion and Concluding Remarksmentioning
Motivated by the substantial instability of the fundamental and high-overtone quasinormal modes, recent developments regarding the notion of black hole pseudospectrum call for numerical results with unprecedented precision.
This work generalizes and improves the matrix method for black hole quasinormal modes to higher orders, specifically aiming at a class of perturbations to the metric featured by discontinuity intimately associated with the quasinormal mode structural instability.
The approach is based on the mock-Chebyshev grid, which guarantees its convergence in the degree of the interpolant.
In practice, solving for black hole quasinormal modes is a formidable task.
The presence of discontinuity poses a further difficulty so that many well-known approaches cannot be employed straightforwardly.
Compared with other viable methods, the modified matrix method is competent in speed and accuracy. 
Therefore, the method serves as a helpful gadget for relevant studies.
“…Feedforward neural network method [100,101]. On the other hand, the method that we will mainly focus on in this review will be the WKB approximation.…”
We give a pedagogical introduction to black holes (BHs) greybody factors (GFs) and quasinormal modes (QNMs) and share the recent developments on those subjects. To this end, we present some particular analytical and approximation techniques for the computations of the GFs and QNMs.We first review the gravitational GFs and show how they are analytically calculated for static and spherically symmetric higher dimensional BHs, consisting the charged BHs and existence of cosmological constant (i.e., de Sitter (dS)/anti-de Sitter (AdS)AdS BHs). The computations performed involve both the low-energy (having real and small frequencies) and the asymptotic (having extremely high frequency of the scattered wave throughout the imaginary axis) cases. A generic method is discussed at low frequencies. This method can be used for all three types of spacetime asymptotics and it is unaffected by the BH's features. For asymptotically dS BHs, GF varies depending on whether the spacetime dimension is even or odd, and is proportional to the ratio of the event and cosmic horizon areas. At asymptotic frequencies, the GFs can be computed by using a matching technique inspired by the monodromy method. In the meantime, we also make a general literature review on the matching technique in a separate section. While the GFs for charged or asymptotically dS BHs are generated by non-trivial functions, the GF for asymptotically AdS BHs is precisely one: pure black-body emission. QNMs, which are solutions to the relevant perturbation equations that satisfy the boundary conditions for purely outgoing (gravitational) waves at spatial infinity and purely ingoing (gravitational) waves at the event horizon, are considered using some particular analytical (like the matching technique) and approximation methods. In this study, our primary focus will be on the bosonic and fermionic GFs and QNMs of various BH and brane geometries and reveal the fingerprints of the invisibles with the radiation spectra to be obtained by the WKB approximation and bounding the Bogoliubov coefficients (together with the Miller-Good transformation) methods.
“…Consequently, other alternative approaches have been proposed for calculating QNM frequencies, including classical and numerical methods such as the Chandrasekhar-Detweiler method, direct integration of the wave equation, the Frobenius series method and its variations, the continued fractions method, and the monodromy technique for highly damped QNMs (for a discussion of these methods see [2,53]). In recent years, novel and alternative computational methods have also been developed to obtain BH QNM frequencies, such as the Borel summation method [54], the Jansen Mathematica package [55] or the use of Neural Networks Methods [56].…”
Section: The Pöschl-teller Potential Methodsmentioning
In this work, we investigate the relationship between the geometrical properties, the photon sphere, the shadow, and the eikonal quasinormal modes of electrically charged black holes in 4D Einstein-Gauss-Bonnet gravity. Quasinormal modes are complex frequency oscillations that are dependent on the geometry of spacetime and have significant applications in studying black hole properties and testing alternative theories of gravity. Here, we focus on the eikonal limit for high frequency quasinormal modes and their connection to the black holes geometric characteristics. To study the photon sphere, quasinormal modes, and black hole shadow, we employ various techniques such as the WKB method in various orders of approximation, the Poschl-Teller potential method, and Churilova's analytical formulas. Our results indicate that the real part of the eikonal quasinormal mode frequencies of test fields are linked to the unstable circular null geodesic and are correlated with the shadow radius for an Charged Einstein-Gauss-Bonnet 4D black hole. Furthermore, we found that the real part of quasinormal modes, the photon sphere and shadow radius have a lower value for charged black holes in 4D Einstein-Gauss-Bonnet gravity compared to black holes without electric charge and those of static black holes in general relativity. Additionally, we explore various analytical formulas for the photon spheres and shadows, deducing an Churilova's approximate formula for the black hole shadow radius of the Charged Einstein-Gauss-Bonnet 4D black hole, which arises from its connection with the eikonal quasinormal modes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.