Vanessa Graber (Institute of Space Sciences ICE-CSIC) Barcelona - Spain, November 13, 2023

Although about a billion neutron stars exist in our Milky Way, observational constraints limit us to only observing a few thousand. Pulsar population synthesis bridges this gap by simulating synthetic populations and comparing these to the observed sample of radio pulsars to constrain uncertain neutron-star physics. In this talk, I will explore the possibility of using simulation-based inference based on artificial neural networks to estimate the parameters governing the magnetic and rotational properties of isolated Galactic radio pulsars. For this purpose, we have developed a flexible population-synthesis pipeline to simulate the neutron stars' dynamical and magneto-rotational evolution as well as their radio emission and incorporated selection biases of typical radio surveys. We subsequently use this framework to generate an extensive database of synthetic pulsar populations to train and validate a mixture-density network to recover the posterior distribution of those parameters that govern the neutron stars' properties at birth. I will conclude by presenting our new inference results for the observed pulsar population.