Speaker
Description
We present, to the best of our knowledge, the first application of Physics-Informed Neural Networks (PINNs) to the neutron star equation-of-state (EOS) inverse problem. Two interacting networks, one representing the EOS $P(\varepsilon)$ as a continuous, non-parametric function and the other solving the Tolman–Oppenheimer–Volkoff (TOV) equations, are trained jointly on NICER X-ray timing posteriors and pulsar mass measurements. The TOV equations are incorporated as a mean-square ODE residual enforced through automatic differentiation at every training step, rooted in the Neural Differential Equation framework. The inferred EOS satisfies nuclear saturation properties, causality, and perturbative QCD bounds simultaneously; χEFT consistency at $1-2 \rho_0$ emerges without explicit enforcement, providing a non-trivial self-consistency check. Across $N=15$ independent training runs, we obtain a neutron star maximum mass of $M_{\max}=2.06^{+0.07}_{-0.09} M_\odot$, and a radius and tidal deformability for a $1.4 M_\odot$ star of $R_{1.4}=12.85^{+0.03}_{-0.06} \mathrm{km}$ and $\Lambda_{1.4}=684$, respectively, with $68\%$ CI, in agreement with recent Bayesian analyses. Most interestingly, the speed of sound exhibits a reproducible softening at $2-4 \rho_0$, consistent with a possible quark–hadron crossover.
| Keywords | Neutron stars, Physics-Informed Neural Networks, Equation of State. |
|---|---|
| Proceedings Publication Intent | No |
| Student Award Eligibility | yes |
| Financial Support Request | no |