19th International Conference on QCD in Extreme Conditions (XQCD 2023)

Rebuilding Neutron Star EoSs from Observations with Deep Learning

26 Jul 2023, 16:35

25m

Auditorium C.1 (Department of Physics (University of Coimbra))

Talk Parallel session A

Speaker

Lingxiao Wang (Frankfurt Institute for Advanced Studies (FIAS))

Description

We present a novel deep learning approach to optimize the equation of state (EoS) for probing neutron star observables. By leveraging an automatic differentiation framework, our method solves inverse problems and achieves accurate EoS optimization. Through training a neural network on a comprehensive dataset, we develop a predictive EoS model that yields precise relationships between pressure, speed of sound, and mass density. Our results align with conventional approaches and are consistent with the observed tidal deformability from the gravitational wave event, GW170817.

Presentation materials

XQCD_slides_lingxiao_20230726.pdf