Pith. sign in

REVIEW 4 major objections 6 minor 55 references

Neural-Spectral Discovery of Rotating Black Holes Beyond General Relativity

T0 review · 4 major / 6 minor · reviewed 2026-07-09 · glm-5.2

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Neural-spectral method finds first rotating black holes in cubic Lovelock gravity

desk verdict First rotating solutions in cubic Lovelock gravity via a PINN-spectral hybrid; the method works but the new solutions lack an independent physical consistency check. read the letter →

arxiv 2607.07418 v1 pith:XKY7C52J submitted 2026-07-08 gr-qc hep-th

Felipe Agurto-Sepúlveda , Marcelo Oyarzo , Anxo Biasi , Devansh Agarwal , Ethan Tregidga , James F. Steiner , Jose D. Edelstein , Gaston Giribet
show 1 more author
Cecilia Garraffo
This is my paper · ORCID
classification gr-qchep-th PACS 04.70.Bw04.20.Jb
keywords blackrotatingequationsmethodsolutionsconstantscouplingdifferential
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces Akribeia, a hybrid computational framework that pairs physics-informed neural networks (PINNs) with pseudo-spectral Chebyshev refinement using extended-precision arithmetic. The method is designed to solve highly non-linear boundary-value problems in gravitational physics without requiring perturbative seeds or parameter continuation. The authors apply Akribeia to Lovelock gravity, a higher-curvature generalization of Einstein's theory, and construct the first fully non-linear rotating black hole solutions in seven-dimensional cubic Lovelock gravity for arbitrary values of the gravitational couplings and angular momenta. The central claim is that the PINN stage, trained from random initialization, can locate the correct solution branch in a rugged landscape where classical Newton-type solvers fail for lack of a good initial guess, and that the subsequent spectral stage then certifies the solution to residuals far below machine precision. The paper validates the method by recovering known five-dimensional Einstein-Gauss-Bonnet rotating solutions and the exact Myers-Perry metric in the General Relativity limit, achieving relative errors as small as 10^{-160}. The new cubic Lovelock solutions reveal that higher-curvature corrections enhance frame dragging and deform the horizon geometry, with a critical coupling marking a transition from oblate to prolate horizon shapes that may signal stability against certain instabilities. The authors argue that this parametric, continuous family of solutions is exactly what is needed to build theoretical templates for horizon-scale imaging and gravitational-wave ringdown observations testing deviations from the Kerr metric.

What carries the argument

The key machinery is the two-stage Akribeia pipeline. Stage one is a PINN with Swish activations, trained via Adam with learning-rate warm-up followed by L-BFGS, using strong (hard-enforced) boundary conditions via network reparameterization and automatic differentiation for all field-equation residuals. Stage two is a pseudo-spectral Chebyshev method with Gauss-Newton optimization and up to 300-digit extended-precision arithmetic that refines the PINN output to certified accuracy. The gravitational ansatz exploits SU(n) symmetry in the equal-angular-momenta sector of odd-dimensional Lovelock gravity, reducing the problem to cohomogeneity-1 (a system of ODEs in the radial coordinate).

What would settle it

A known exact or independently computed numerical solution for rotating cubic Lovelock black holes at specific coupling values that disagrees with the Akribeia output would falsify the claim. More immediately, if the PINN were shown to converge to a different solution branch than the one found by a trusted continuation method in a regime where both are applicable (e.g., the EGB validation case), the reliability of the seed-generation stage would be undermined.

Watch

Extended reading notes

Core claim

The paper's central object is a certified neural-field solution: a continuous, globally defined metric function parametrized by coupling constants, produced by the Akribeia pipeline. The core discovery is that this pipeline can find, from random initialization, the correct physical solution branch for rotating black holes in seven-dimensional cubic Lovelock gravity, a system governed by a highly non-linear, non-perturbative set of coupled ODEs where no fully non-linear rotating solution was previously known. The PINN identifies an approximate solution valley; the pseudo-spectral Chebyshev solver with extended-precision arithmetic then drives residuals to extreme precision, yielding a Chebysh

Load-bearing premise

The method assumes that the PINN, trained on collocation points with residuals reduced to roughly 10^{-3}, reliably converges to the correct physical solution branch rather than a spurious local minimum. If the PINN lands on a mathematically valid but physically incorrect branch, the spectral refiner would converge to a wrong solution, and the paper does not provide a rigorous proof that the desired branch is always selected.

Editorial extensions

If this is right

  • The method can be systematically extended to higher Lovelock orders (quartic and above), to asymptotically Anti-de Sitter backgrounds, and to other beyond-GR theories admitting a cohomogeneity-1 reduction, potentially producing a broad catalogue of non-Kerr rotating black hole solutions.
  • Certified parametric solution families could be folded into observational pipelines for the Event Horizon Telescope, LISA, and X-ray missions to quantify or break degeneracies between black-hole spin and higher-curvature couplings in measurements of shadows, photon rings, and ringdown spectra.
  • The enhanced frame dragging (w(r) increasing with coupling) and the horizon-shape transition (oblate to prolate at a critical coupling) are specific, testable predictions for how higher-curvature gravity modifies strong-field observables relative to the Kerr metric.
  • Linearized perturbation analysis around these new solutions could determine whether the prolate-horizon regime suppresses Gregory-Laflamme-type instabilities that affect ultra-spinning black holes in lower-curvature theories.
  • The Akribeia framework is not limited to gravitational physics; the paper notes its applicability to other non-linear PDE and ODE systems, suggesting a general-purpose tool for discovering solution branches in stiff boundary-value problems across physics.
  • Adaptation to full PDE systems (beyond the cohomogeneity-1 ODE reduction) would enable construction of rotating solutions with unequal angular momenta, removing the equal-spin symmetry assumption and broadening the astrophysical relevance of the catalogue.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. This Letter introduces Akribeia, a hybrid framework combining physics-informed neural networks (PINNs) with pseudo-spectral Chebyshev refinement using extended-precision arithmetic, and applies it to construct rotating black hole solutions in Lovelock gravity. The method is validated by recovering the Myers-Perry solution in the GR limit (relative error ~10^{-160}) and reproducing known 5D Einstein-Gauss-Bonnet (EGB) rotating solutions from Brihaye et al. [27]. The paper's central new result is the construction of fully non-linear rotating black hole solutions in 7D cubic Lovelock gravity for arbitrary couplings and equal angular momenta, which to the authors' knowledge are the first such solutions reported.

Significance. The problem of finding stationary rotating black hole solutions in higher-curvature gravity is long-standing and important. The hybrid neural-spectral approach is methodologically interesting: the PINN stage provides a seed-free initialization that bypasses the continuation/perturbative-seed bottleneck of standard Newton-type solvers, and the spectral stage delivers certified high-precision solutions. The recovery of Myers-Perry to 10^{-160} and the agreement with [27] in the EGB sector are non-trivial validations. If the cubic Lovelock solutions are correct, this fills a genuine gap in the literature. The parametric neural-field representation is a practical strength for future template construction.

major comments (4)
  1. The central new claim — rotating black holes in 7D cubic Lovelock gravity — is validated only by (i) small PDE residuals and (ii) qualitative similarity to the EGB case. However, small residuals are the quantity the method is explicitly designed to minimize, so this check is partially circular. The paper does not report the specific residual level achieved for the cubic solutions (the 10^{-160} figure is only for Myers-Perry recovery). More importantly, no independent physical consistency condition is verified. The first law of black hole mechanics (M = T_H S + Ω_H J) and the Smarr relation provide non-trivial, independent tests because the ADM mass, Wald entropy, Hawking temperature, and angular momentum are extracted from different structural parts of the metric and are not directly imposed as constraints in the ODE boundary-value problem. If the PINN-spectral pipeline converged to asp
  2. ur
  3. ious or unphysical branch, residuals could still be small while the first law fails. The paper itself notes that the cubic case required a deeper network (4×64 vs 3×20), suggesting increased solution-landscape complexity where branch misidentification is more likely. Verification of the first law (or at minimum, a convergence study showing stability of the solution under increasing collocation density and network depth) is needed to support the central claim. See the discussion around Fig. 3 and the paragraph beginning 'Due to the increased difficulty introduced by the cubic term.'
  4. No convergence diagnostics are reported for the cubic solutions. The paper states that residuals are reduced 'far below machine double-precision' but does not provide a quantitative figure, nor does it show how residuals scale with the number of Chebyshev collocation points or network architecture. For the EGB validation (Fig. 1), the comparison with [27] serves as an implicit convergence check, but no such external benchmark exists for the cubic case. Without this information, the reader cannot assess whether the reported solutions are fully converged. A table or plot showing residual levels as a function of spectral resolution for representative cubic-coupling values would address this.
minor comments (6)
  1. The name 'Akribeia' is introduced without explanation of its etymological origin; a brief note would help readers.
  2. In the paragraph describing the training strategy (four stages), the transition from PINN residuals ~10^{-3} to the spectral stage is abrupt. A sentence clarifying what residual threshold triggers the handoff to the spectral solver would improve reproducibility.
  3. Fig. 3 caption: it would help to explicitly state which curves correspond to which β values, or at least indicate the direction of increasing β in the gradient bar.
  4. The asymptotic boundary conditions listed (b(r) → 1 - O(r^{-2n+2}), etc.) use a notation that could be confused with O(r^{-(2n-2)}); clarifying the exponent convention would avoid ambiguity.
  5. Reference [48] is cited as motivation for two-scale systems but appears to be an arXiv preprint (2512.04083); please verify it is appropriately cited.
  6. The statement 'our hybrid approach successfully passes the initialization sensitivities of spectral solvers' (page 5) is vague; specifying what sensitivities were tested would strengthen this claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for a careful and constructive report. The referee correctly identifies that the central new claim — rotating black holes in 7D cubic Lovelock gravity — requires stronger validation than what appears in the current manuscript. We agree with both major comments and will revise accordingly.

read point-by-point responses

  1. Referee: The central new claim is validated only by (i) small PDE residuals and (ii) qualitative similarity to the EGB case. Small residuals are the quantity the method is designed to minimize, so this check is partially circular. No specific residual level is reported for the cubic solutions. No independent physical consistency condition (first law, Smarr relation) is verified. The deeper network architecture for the cubic case suggests increased solution-landscape complexity where branch misidentification is more likely.

    Authors: The referee raises a valid and important point. We agree that residual minimization alone is not a fully independent check, and that the first law of black hole mechanics (M = T_H S + Ω_H J) and the Smarr relation provide non-trivial, structurally independent tests because the ADM mass, Wald entropy, Hawking temperature, and angular momentum are extracted from different parts of the metric and are not directly imposed as constraints in the boundary-value problem. We will verify the first law and Smarr relation for representative cubic-coupling values in the revised manuscript. We note that the Wald entropy in cubic Lovelock gravity involves a specific combination of curvature invariants evaluated at the horizon, and the ADM mass requires asymptotic expansion of the metric functions — both of which are computed from the solution but are not among the field equations being minimized. This makes the first law a genuinely independent diagnostic. We will report the level of agreement (or discrepancy) for several points in parameter space, including cases with non-vanishing cubic coupling β. If the first law is satisfied to a level consistent with the spectral residuals, this will provide the independent physical consistency check the referee rightly requests. We will also report the specific residual levels achieved for the cubic solutions, which were not stated in the original manuscript. revision: yes

  2. Referee: No convergence diagnostics are reported for the cubic solutions. The paper states residuals are reduced 'far below machine double-precision' but does not provide a quantitative figure, nor does it show how residuals scale with the number of Chebyshev collocation points or network architecture. For the EGB validation, comparison with [27] serves as an implicit convergence check, but no such external benchmark exists for the cubic case. A table or plot showing residual levels as a function of spectral resolution for representative cubic-coupling values would address this.

    Authors: This is correct and we will address it. We will add a convergence study showing how the spectral residuals scale with the number of Chebyshev collocation points for representative cubic-coupling values (e.g., β = 0.1, 0.3, 0.5 with α = 0.5 fixed, matching the parameters of Fig. 3). We will also report the quantitative residual levels achieved after spectral refinement for these configurations. Additionally, we will include a brief study of solution stability under variation of network architecture (depth and width) to demonstrate that the reported solutions are not artifacts of a particular network configuration. This will take the form of a table or figure in the revised manuscript or Supplemental Material. revision: yes

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: solutions are derived from field equations, not fitted to data; validation is against external benchmarks.

full rationale

The paper constructs rotating black hole solutions by minimizing PDE residuals of the Lovelock field equations using a PINN-spectral hybrid method. The solutions are not fitted to observational data or to the target results; they are obtained by solving the field equations from random initialization. Validation is performed against external benchmarks: the exact Myers-Perry solution (GR limit) and the numerical EGB solutions of Brihaye et al. [27], with which the authors report agreement. The 10^{-160} residual figure is for Myers-Perry recovery, not for the new cubic solutions, but this is a precision benchmark, not a circular definition. The new cubic Lovelock solutions in 7D are derived from the field equations with boundary conditions, not from fitting to known results. The qualitative similarity check between cubic and quadratic solutions is a heuristic consistency argument, not a load-bearing circular definition. The skeptic's concern about absence of first-law/Smarr verification is a correctness/completeness concern, not a circularity issue: the solutions are not defined in terms of the first law. No step in the derivation chain reduces to its own inputs by construction. The only minor concern is that residual minimization is both the method and the validation metric, but this is inherent to any numerical PDE solver and does not constitute circularity in the sense of the result being equivalent to the input by definition.

Assumptions & free parameters 5 free parameters · 3 assumptions · 1 invented entities

The free parameters are physical coupling constants and scale parameters, not fitted constants introduced ad hoc. The main ad hoc assumption is the convergence of the hybrid method to the correct solution branch.

free parameters (5)
  • alpha = varied in [0,3] for EGB; fixed at 0.5 for cubic
    Quadratic Lovelock coupling constant, treated as an input parameter for the solution family.
  • beta = varied in [0,0.5] for cubic
    Cubic Lovelock coupling constant, treated as an input parameter for the solution family.
  • r_H = 1
    Horizon radius, fixed as a scale parameter.
  • Omega_H = 0.33
    Angular velocity at the horizon, fixed for the presented solutions.
  • Neural Network Weights (W) = optimized
    Weights and biases of the PINN, optimized to minimize the PDE residuals.
assumptions (3)
  • domain assumption Lovelock gravity action (Eq. 1) is the correct effective gravitational theory.
    The field equations solved are derived from this action.
  • domain assumption The co-homogeneity-1 metric ansatz (Eq. 4) with equal angular momenta captures the relevant physics.
    The reduction to ODEs relies on this symmetry assumption.
  • ad hoc to paper The PINN provides a sufficiently accurate initialization for the spectral solver to converge to the correct physical solution branch.
    This is the core assumption enabling the hybrid method; if false, the spectral solver could converge to a spurious solution.
invented entities (1)
  • Akribeia independent evidence
    purpose: The hybrid neural-spectral framework for solving non-linear PDEs.
    The method is validated against known exact and numerical solutions, demonstrating its efficacy.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Neural-Spectral Discovery of Rotating Black Holes Beyond General Relativity." pith.science (2026). https://pith.science/paper/XKY7C52J

@misc{pith2026260707418,
  author       = {Pith},
  title        = {Pith review of: Neural-Spectral Discovery of Rotating Black Holes Beyond General Relativity},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XKY7C52J}},
  note         = {Machine review of arXiv:2607.07418}
}
read the original abstract

Finding rotating black hole solutions in higher-curvature theories of gravity is a problem of fundamental importance. Virtually every approach to reconcile gravity with quantum mechanics predicts corrections to the Einstein-Hilbert action, yet no systematic solution-generating method exists for the stationary sector. We close this gap with {\sc Akribeia}, a novel hybrid framework that pairs physics-informed neural networks with a pseudo-spectral refinement step, yielding certified neural-field rotating black hole solutions -- continuous, globally defined functions, parametric in the coupling constants -- whose residuals against the field equations are verified to extreme precision. We apply the method to theories quadratic and cubic in the curvature and construct, for the first time, families of rotating black holes featuring multiple non-vanishing angular momenta, parametric in the new coupling constants. After validating against previously known five-dimensional spacetimes, we present new solutions in scenarios leading to a highly non-linear/non-perturbative coupled system of ordinary differential equations. Our method can be systematically adapted to other setups involving partial differential equations as well.

Figures

Figures reproduced from arXiv: 2607.07418 by the authors.

Figure 1
Figure 1. FIG. 1: Rotating solution with equal angular momenta for [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 3
Figure 3. FIG. 3: Rotating black hole solutions with equal angular [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 55 canonical work pages

  1. [27]

    Introduction to numerical continuation methods

    E. L. Allgower and K. Georg, “Introduction to numerical continuation methods” (Classics in Applied Mathematics, Vol.45), SocietyforIndustrialandApplied Mathematics, 2003

    work page 2003

  2. [1]

    The second stage consists of keeping the learning rate stable at10 −3 to broadly explore the highly non-linear solution landscape

    with a learning rate warm-up, which starts at10−6 and increases linearly to reach10−3. The second stage consists of keeping the learning rate stable at10 −3 to broadly explore the highly non-linear solution landscape. Once a promising valley is identified, the third stage refines the search using an adaptive learning rate decay based on the validation los...

    work page

  3. [2]

    In contrast, the PINN training procedure implemented here can construct, from a random initialization, a continuous family of solutions parametrized byα∈[0,3]

    (ΩH ̸= 0,α= 0) or the Boulware-Deser solution [10] (ΩH = 0,α̸= 0) as seeds, and then gradually increase the coupling constantαand/orΩ H ∼ O(a)until the desired configuration is reached. In contrast, the PINN training procedure implemented here can construct, from a random initialization, a continuous family of solutions parametrized byα∈[0,3]. Figure 1 sh...

    work page 2021

  4. [3]

    Renormalization of Higher Derivative Quantum Gravity,

    K. S. Stelle, “Renormalization of Higher Derivative Quantum Gravity,” Phys. Rev. D16, 953 (1977)

    work page 1977

  5. [4]

    Superstring Modifications of Einstein’s Equations,

    D. J. Gross and E. Witten, “Superstring Modifications of Einstein’s Equations,” Nucl. Phys. B277, 1 (1986)

    work page 1986

  6. [5]

    Curvature Squared Terms and String Theories,

    B. Zwiebach, “Curvature Squared Terms and String Theories,” Phys. Lett. B156, 315 (1985)

    work page 1985

  7. [6]

    First Sagittarius A* Event Horizon Telescope Results. VI. Testing the Black Hole Metric,

    K. Akiyamaet al.[Event Horizon Telescope], “First Sagittarius A* Event Horizon Telescope Results. VI. Testing the Black Hole Metric,” Astrophys. J. Lett.930, L17 (2022)

    work page 2022

  8. [7]

    GW250114: Testing Hawking’s Area Law and the Kerr Nature of Black Holes,

    A. G. Abacet al.[LIGO Scientific, Virgo and KAGRA], “GW250114: Testing Hawking’s Area Law and the Kerr Nature of Black Holes,” Phys. Rev. Lett.135, 111403 (2025)

    work page 2025

Show all 55 references
  1. [8]

    Black hole spectroscopy beyond Kerr: Agnostic and theory-based tests with next- generation interferometers,

    A. Maselli, S. Yi, L. Pierini, V. Vellucci, L. Reali, L. Gualtieri and E. Berti, “Black hole spectroscopy beyond Kerr: Agnostic and theory-based tests with next- generation interferometers,” Phys. Rev. D109, 064060 (2024)

    2024

  2. [9]

    Higher-derivative corrections to the Kerr quasinormal mode spectrum,

    P. A. Cano, L. Capuano, N. Franchini, S. Maenaut and S. H. Völkel, “Higher-derivative corrections to the Kerr quasinormal mode spectrum,” Phys. Rev. D110, 124057 (2024)

    2024

  3. [10]

    A Remarkable property of the Riemann- Christoffel tensor in four dimensions,

    C. Lanczos, “A Remarkable property of the Riemann- Christoffel tensor in four dimensions,” Annals Math.39, 842 (1938)

    1938

  4. [11]

    The Einstein tensor and its generalizations,

    D. Lovelock, “The Einstein tensor and its generalizations,” J. Math. Phys.12, 498 (1971). 7

    1971

  5. [12]

    String Generated Gravity Models,

    D. G. Boulware and S. Deser, “String Generated Gravity Models,” Phys. Rev. Lett.55, 2656 (1985)

    1985

  6. [13]

    Classical Gravity with Higher Derivatives,

    K. S. Stelle, “Classical Gravity with Higher Derivatives,” Gen. Rel. Grav.9, 353 (1978)

    1978

  7. [14]

    Causality in AdS/CFT and Lovelock theory,

    X. O. Camanho and J. D. Edelstein, “Causality in AdS/CFT and Lovelock theory,” JHEP06, 099 (2010)

    2010

  8. [15]

    Lovelock theory and the AdS/CFT correspondence,

    X. O. Camanho, J. D. Edelstein and J. M. Sánchez De Santos, “Lovelock theory and the AdS/CFT correspondence,” Gen. Rel. Grav.46, 1637 (2014)

    2014

  9. [16]

    A New type of phase transition in gravitational theories,

    X. O. Camanho, J. D. Edelstein, G. Giribet and A. Gomberoff, “A New type of phase transition in gravitational theories,” Phys. Rev. D86, 124048 (2012)

    2012

  10. [17]

    Generalized phase transitions in Lovelock gravity,

    X. O. Camanho, J. D. Edelstein, G. Giribet and A. Gomberoff, “Generalized phase transitions in Lovelock gravity,” Phys. Rev. D90, 064028 (2014)

    2014

  11. [18]

    Critical Points of D-Dimensional Extended Gravities,

    S. Deser, H. Liu, H. Lu, C. N. Pope, T. C. Sisman and B. Tekin, “Critical Points of D-Dimensional Extended Gravities,” Phys. Rev. D83, 061502 (2011)

    2011

  12. [19]

    The Lovelock Black Holes,

    C. Garraffo and G. Giribet, “The Lovelock Black Holes,” Mod. Phys. Lett. A23, 1801 (2008)

    2008

  13. [20]

    A Lovelock black hole bestiary,

    X. O. Camanho and J. D. Edelstein, “A Lovelock black hole bestiary,” Class. Quant. Grav.30, 035009 (2013)

    2013

  14. [21]

    Cosmic censorship in Lovelock theory,

    X. O. Camanho and J. D. Edelstein, “Cosmic censorship in Lovelock theory,” JHEP11, 151 (2013)

    2013

  15. [22]

    Kerr-Schild ansatz in Einstein-Gauss-Bonnet gravity: An exact vacuum solution in five dimensions,

    A. Anabalon, N. Deruelle, Y. Morisawa, J. Oliva, M. Sasaki, D. Tempo and R. Troncoso, “Kerr-Schild ansatz in Einstein-Gauss-Bonnet gravity: An exact vacuum solution in five dimensions,” Class. Quant. Grav. 26, 065002 (2009)

    2009

  16. [23]

    Kerr-Schild Ansatz in Lovelock Gravity,

    B. Ett and D. Kastor, “Kerr-Schild Ansatz in Lovelock Gravity,” JHEP04, 109 (2011)

    2011

  17. [24]

    Gravitational field of a spinning mass as an example of algebraically special metrics,

    R. P. Kerr, “Gravitational field of a spinning mass as an example of algebraically special metrics,” Phys. Rev. Lett.11, 237-238 (1963)

    1963

  18. [25]

    Five-dimensional rotating black holes in Einstein-Gauss-Bonnet theory,

    Y. Brihaye and E. Radu, “Five-dimensional rotating black holes in Einstein-Gauss-Bonnet theory,” Phys. Lett. B661, 167 (2008)

    2008

  19. [26]

    Slowly Rotating Charged Gauss-Bonnet Black holes in AdS Spaces,

    H. C. Kim and R. G. Cai, “Slowly Rotating Charged Gauss-Bonnet Black holes in AdS Spaces,” Phys. Rev. D77, 024045 (2008)

    2008

  20. [28]

    Nonlinearly preconditioned inexact Newton algorithms,

    X. C. Cai and D. E. Keyes, “Nonlinearly preconditioned inexact Newton algorithms,” SIAM Journal on Scientific Computing24, 183 (2002)

    2002

  21. [29]

    Rotating black holes with equal-magnitude angular momenta in d=5 Einstein-Gauss-Bonnet theory,

    Y. Brihaye, B. Kleihaus, J. Kunz and E. Radu, “Rotating black holes with equal-magnitude angular momenta in d=5 Einstein-Gauss-Bonnet theory,” JHEP11(2010), 098

    2010

  22. [30]

    Spinning black strings in five-dimensional Einstein-Gauss-Bonnet gravity,

    B.Kleihaus, J.Kunz, E.RaduandB.Subagyo, “Spinning black strings in five-dimensional Einstein-Gauss-Bonnet gravity,” Phys. Lett. B713, 110 (2012)

    2012

  23. [31]

    Rotating black holes in Einstein-Dilaton-Gauss-Bonnet gravity with finite coupling,

    A. Maselli, P. Pani, L. Gualtieri and V. Ferrari, “Rotating black holes in Einstein-Dilaton-Gauss-Bonnet gravity with finite coupling,” Phys. Rev. D92, no.8, 083014 (2015)

    2015

  24. [32]

    Spinning black holes in Einstein-Gauss-Bonnet-dilaton theory: Nonperturbative solutions,

    B. Kleihaus, J. Kunz, S. Mojica and E. Radu, “Spinning black holes in Einstein-Gauss-Bonnet-dilaton theory: Nonperturbative solutions,” Phys. Rev. D93, 044047 (2016)

    2016

  25. [33]

    Physics- informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partialdifferentialequations,

    M. Raissi, P. Perdikaris and G. E. Karniadakis, “Physics- informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partialdifferentialequations,” J.Comput.Phys.378, 686 (2019)

    2019

  26. [34]

    Learning anddiscoveringmultiplesolutionsusingphysics-informed neural networks with random initialization and deep ensemble,

    Z. Zou, Z. Wang and G. Em Karniadakis, “Learning anddiscoveringmultiplesolutionsusingphysics-informed neural networks with random initialization and deep ensemble,” Proc. A481, 2325 (2025)

    2025

  27. [35]

    Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations,

    M. Raissi, A. Yazdani and G. E. Karniadakis, “Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations,” Science367, 1026 (2020)

    2020

  28. [36]

    Asymptotic self-similar blow-up profile for three-dimensional axisymmetric Euler equations using neural networks,

    Y. Wang, C. Y. Lai, J. Gómez-Serrano and T. Buckmaster, “Asymptotic self-similar blow-up profile for three-dimensional axisymmetric Euler equations using neural networks,” Phys. Rev. Lett.130, 244002 (2023)

    2023

  29. [37]

    Discovery of unstable singularities,

    Y. Wanget al., “Discovery of unstable singularities,” arXiv:2509.14185

  30. [38]

    Solving the Teukolsky equation with physics-informed neural networks,

    R. Luna, J. Calderón Bustillo, J. J. S. Martínez, A. Torres-Forné and J. A. Font, “Solving the Teukolsky equation with physics-informed neural networks,” Phys. Rev. D107, 064025 (2023)

    2023

  31. [39]

    Using physics- informed neural networks to compute quasinormal modes,

    A. S. Cornell, A. Ncube and G. Harmsen, “Using physics- informed neural networks to compute quasinormal modes,” Phys. Rev. D106, 124047 (2022)

    2022

  32. [40]

    Teukolsky by design : A hybrid spectral-PINN solver for Kerr quasinormal modes,

    A. M. Pombo and L. Pizzuti, “Teukolsky by design : A hybrid spectral-PINN solver for Kerr quasinormal modes,” JCAP03, 009 (2026)

    2026

  33. [41]

    Addressing the gravitational collapse of a massless scalar field with physics-informed neural networks,

    A. Ferrer-Sánchez, N. Villanueva-Espinosa, C. Hernani- Morales, R. Ruiz de Austri-Bazan, J. A. Font, J. David Martín-Guerrero and M. W. Choptuik, “Addressing the gravitational collapse of a massless scalar field with physics-informed neural networks,” Mach. Learn.: Sci. Techno...

    2026

  34. [42]

    NeuroSEM: A hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements,

    K. Shukla, Z. Zou, C. H. Chan, A. Pandey, Z. Wang and G. E. Karniadakis, “NeuroSEM: A hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements,” Comput. Methods Appl. Mech. Engrg.433, 117498 (2025)

    2025

  35. [43]

    Slowly rotating charged black holes in anti-de Sitter third order Lovelock gravity,

    R. Yue, D. Zou, T. Yu, P. Li and Z. Yang, “Slowly rotating charged black holes in anti-de Sitter third order Lovelock gravity,” Gen. Rel. Grav.43, 2103 (2011)

    2011

  36. [44]

    Slowly rotating black holes in Einsteinian cubic gravity,

    C. Adair, P. Bueno, P. A. Cano, R. A. Hennigar and R. B. Mann, “Slowly rotating black holes in Einsteinian cubic gravity,” Phys. Rev. D102, 084001 (2020)

    2020

  37. [45]

    Adam: A Method for Stochastic Optimization,

    D. P. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” [arXiv:1412.6980 [cs.LG]]

    arXiv

  38. [46]

    On the limited memory BFGS method for large scale optimization,

    D. C. Liu and J. Nocedal, “On the limited memory BFGS method for large scale optimization,” Mathematical Programming45, 503 (1989)

    1989

  39. [47]

    Black Holes in Higher Dimensional Space-Times,

    R. C. Myers and M. J. Perry, “Black Holes in Higher Dimensional Space-Times,” Annals Phys.172, 304 (1986)

    1986

  40. [48]

    Gravitational perturbations of higher dimensional rotating black holes: Tensor perturbations,

    H. K. Kunduri, J. Lucietti and H. S. Reall, “Gravitational perturbations of higher dimensional rotating black holes: Tensor perturbations,” Phys. Rev. D74, 084021 (2006)

    2006

  41. [49]

    Instability of ultra- spinning black holes,

    R. Emparan and R. C. Myers, “Instability of ultra- spinning black holes,” JHEP09, 025 (2003)

    2003

  42. [50]

    Screening of dipolar emission in two-scale Gauss-Bonnet gravity,

    F. Thaalba, L. Gualtieri, T. P. Sotiriou and E. Trincherini, “Screening of dipolar emission in two-scale Gauss-Bonnet gravity,” arXiv:2512.04083

  43. [51]

    Black Hole Spin via Continuum Fitting and the Role of Spin in Powering Transient Jets,

    J. E. McClintock, R. Narayan and J. F. Steiner, “Black Hole Spin via Continuum Fitting and the Role of Spin in Powering Transient Jets,” Space Sci. Rev.183, 295 (2014). 8

    2014

  44. [52]

    ObservationalConstraintsonBlackHole Spin,

    C.S.Reynolds, “ObservationalConstraintsonBlackHole Spin,” Ann. Rev. Astron. Astrophys.59, 117 (2021)

    2021

  45. [53]

    Polarized x-rays constrain the disk-jet geometry in the black hole x-ray binary Cygnus X-1,

    H. Krawczynskiet al., “Polarized x-rays constrain the disk-jet geometry in the black hole x-ray binary Cygnus X-1,” Science378, 650 (2022)

    2022

  46. [54]

    A review of quasi-periodic oscillations from black hole X-ray binaries: observation and theory,

    A. Ingram and S. Motta, “A review of quasi-periodic oscillations from black hole X-ray binaries: observation and theory,” New Astron. Rev.85, 101524 (2019)

    2019

  47. [55]

    Testing black hole candidates with electromagnetic radiation,

    C. Bambi, “Testing black hole candidates with electromagnetic radiation,” Rev. Mod. Phys.89, 025001 (2017)

    2017

Pith tools

Reviewed July 9, 2026 · model on record in the stance chip above.