Physics-Informed Deep Learning for Three Dimensional Black Holes

Document Type : Research Paper

Authors

1 Instituut-Lorentz for Theoretical Physics, ITP, Leiden University

2 Faculty of Physics, Semnan Universiy, Semnan, Iran, 35131-19111

Abstract

In this paper, we have designed an artificial neural network architecture to produce metric field of planar BTZ and quintessence black holes applying a data-driven approach andleveraging holography principle (according to AdS/DL (Anti de Sitter/ Deep Learning) correspondence given by [1]). Data has been collected by choosing minimally coupled massive scalar field with quantum fluctuations and we try to process two emergent and ground-truth metrics versus the holographic parameter which plays the role of depth of the neural network. Loss or error function which shows rate of deviation of these two metrics in presence of penalty regularization term reaches to its minimum value when values of the learning rate approach to the observed steepest gradient point. Values of the regularization or penalty term of the quantum scalar field has critical role to matching this two mentioned metric. Also, we design an algorithm which helps us to find optimum value for learning parameter and finally, we understand that loss function convergence heavily depends on the number of epochs and learning rate.

Keywords


[1] Hashimoto, K., Sugishita, S., Tanaka A., & Tomiya, A. 2018, Deep learning and the AdS/CFT correspondence, Phys. Rev. D4, 046019.
[2] Hawking, S. 1976, Black Holes and Thermodynamics, Phys. Rev. D13, 191.
[3] Bekenstein, D. 1973, Black Holes and Entropy, Phys. Rev. D7, 2333.
[4] Hooft, G. 1995, Dimensional Reduction in Quantum Gravity.
[5] Susskind, L. 1995, J. Math. Phys., 36, 6377.
[6] Maldacena, J. M. 1998, Int. J. Theor. Phys., 2, 231.
[7] Banados, M., Teitelboim, C., & Zanelli, J. 1992, Phys. Rev. Lett., 69, 1849.
[8] Sa, P., Kleber, A., & Lemos, J. 1996, Class. Quantum Grav., 13, 125.
[9] Son, D., & Starinets, A. 2002, JHEP, 09, 042.
[10] Abdalla, E., de Oliveira, J., & Lima Santos, A. 2012, Phys. Lett. B709, 276.
[11] Kiselev, V. 2003, Class. Quantum Grav., 20, 1187.
[12] Ghaffarnejad, H., Yaraie, E., & Farsam, M. 2018, Int. J. Theor. Phys., 57, 1671.
[13] Ghaffarnejad, H., Yaraie, E., & Farsam, M. 2020, Advances in High Energy Physics, 9529356.
[14] Ghaffarnejad, H., Yaraie, E., & Farsam, M. 2020, Eur. Phys. J. Plus, 135, 179.
[15] Chen, S., Pan, Q., & Jing, J. 2013, Class. Quantum Grav., 14, 145001.
[16] de Oliveira, J., & Fontana, R. 2018, Phys. Rev. D98, 044005.
[17] Banados, M., Teitelboim, C., & Zanelli, J.1992, Phys. Rev. Lett., 69, 1849.
[18] Banados, M., Henneaux, M., Teitelboim, C., & Zanelli, J. 2013, Erratum Phys. Rev. D, 88, 069902.
[19] Matsueda, H., & Suzuki, T. 2017, J. Phys. Soc. Jpn., 86, 104001.
[20] Bircanoglu, C. 2017, www.researchgate.net,  publication: 318588371.
[21] Carleo, G., & et al. 2019, Rev. Mod. Phys., 91, 045002.
[22] Yang, Y., Zhao, J., Zhang, J., Ye, X., & Zhao, G. 2020, The Astron. J., 160, 236.
[23] Christopher, J., & Vanderburg, A. 2018, The Astorn. J., 155, 94.
[24] Karniadakis, G. E., Kevrekidis, I. G., & Yang, L. 2021, Nature Rev. Phys., 3, 422.
[25] Matsuoka, D., Watanabe, S., Sato, K., Kawazoe, S., & et al. 2020, Geophysical Research Letters, 47, 1.
[26] Rosenblatt, F. 1958, Psyc. Rev., 65, 386.
[27] Hinton, G. E., & Salakhutdinov, R. R. 2006, Science, 313, 504.
[28] LeCun, Y., Bengio, Y., & Hinton, G. 2015, Nature, 521, 436.
[29] Rawat, W., & Wang, Z. 2017, Neural Comput., 29, 2352.
[30] Yu, D., & Li, J. 2017, IEEE/CAA J. Autom. Sinica, 4, 396.
[31] Young, T., Hazarika, D., Poria, S., & Cambria, E. 2017, IEEE Comput. Intell. Mag., 13, 55.
[32] Cao, C., & et al. 2018, Genomics Proteomics Bioinform., 16, 17.
[33] Rumelhart, D. E., Hinton, G. E., Williams, R. J. 1986, Nature. 323, 533.
[34] Paszke, A., Gross, S., Chintala, S., & et al. 2017, Curran Associates.
[35] Kingma, D. P., & Ba, J. 2017, Adam: A Method for Stochastic Optimization, arXiv:1412.6980 [cs.LG].
[36] Anzanello, M., & Fogliatto, F. 2011, Int. J. of Indus. Ergo., 41, 573.
[37] Leslie, S. N. 2017, Cyclical learning rates for training Neural Networks, arXiv:1506.01186 [cs.CV].