Mainz 2022 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
HK: Fachverband Physik der Hadronen und Kerne
HK 33: Nuclear Astrophysics II
HK 33.3: Talk
Tuesday, March 29, 2022, 16:45–17:00, HK-H10
Neural network reconstruction of the dense matter equation of state from neutron star observables — •Shriya Soma1, Lingxiao Wang1, Shuzhe Shi2, Horst Stoecker1, and Kai Zhou1 — 1Frankfurt Institute for Advanced Studies, Frankfurt am Main, Germany — 2Stony Brook University, Stony Brook, New York, USA
The equation of state (EoS) of strongly interacting cold and ultra-dense matter still remains a major challenge in the field of nuclear physics. With the advancements in measurements of neutron star masses, radii and tidal deformabilities from electromagnetic and gravitational wave observations, neutron stars play an important role in constraining the EoS. In this work, we present a novel method that exploits deep learning techniques to reconstruct the dense matter EoS from mass-radius (M-R) observations of neutron stars. We employ neural networks (NNs) to represent the EoS in a model-independent way, within the range 1-7.4 times the nuclear saturation density. In an unsupervised manner, we implement the Automatic Differentiation (AD) framework to optimize the EoS, so as to yield an M-R curve that best fits the observations. We demonstrate the rebuilding of an EoS on mock data, i.e., M-R pairs derived from a generated set of polytropic EoSs. We show that it is possible to reconstruct the EoS with reasonable accuracy, using just 12 mock M-R pairs, which is nearly equivalent to the current number of observations. We finally deploy the NNs in the AD scheme on real M-R data, including the recent measurements from NICER, to infer the neutron star EoS and present the results hereof.