Bea, YagoJiménez, Raúl (Jiménez Tellado)Mateos, David (Mateos Solé)Shuheng, LiuProtopapas, PavlosTarancón Álvarez, PedroTejerina Pérez, Pablo2025-07-102025-07-102024-07-101126-6708https://hdl.handle.net/2445/222139Holography relates gravitational theories in five dimensions to four-dimensionalquantum field theories in flat space. Under this map, the equation of state of the field theoryis encoded in the black hole solutions of the gravitational theory. Solving the five-dimensionalEinstein’s equations to determine the equation of state is an algorithmic, direct problem.Determining the gravitational theory that gives rise to a prescribed equation of state is amuch more challenging, inverse problem. We present a novel approach to solve this problembased on physics-informed neural networks. The resulting algorithm is not only data-drivenbut also informed by the physics of the Einstein’s equations. We successfully apply it totheories with crossovers, first- and second-order phase transitions37 p.application/pdfengcc-by (c) Bea, Y. et al., 2024http://creativecommons.org/licenses/by/4.0/GravitacióEquacions de camp d'EinsteinGravitationEinstein field equationsinfo:eu-repo/semantics/article7549792025-07-10info:eu-repo/semantics/openAccess