Inicio: 12:00h
Ponente: Álvaro Sánchez-Monge (University of Cologne)
Lugar: Zoom: https://us02web.zoom.us/j/85044910172?pwd=bXRmeWx3cTdDSzkrMFFJUXI5eE9hQT09 Meeting ID: 850 4491 0172 Passcode: 996691

The giant molecular cloud complex Sagittarius B2 (hereafter, SgrB2) is the most massive region with ongoing star formation in our Galaxy. It is located at a projected distance of only 100 pc along the plane to the Galactic Center and at 8.5 kpc from the Sun. The whole complex contains a total gas mass of 10^7 Msun, with the main sites of active star formation corresponding to the hot molecular cores SgrB2(N) and SgrB2(M), which are located at the center of the complex. They contain more than 50 high-mass stars with spectral types ranging from O5 to B0, and constitute one of the best laboratories for the search of new chemical species in the Universe.
In the last years, we have pursued a large project to characterize the structure of SgrB2 combining observations at different wavelengths and sensitive to all the scales (ranging from the large envelope that extends about 40 pc down to a few hundred AU), with 3D radiative transfer modelling. Some of the main results reveal (i) distributed (high-mass) star formation happening throughout the whole SgrB2 envelope and not only in the central hot molecular cores, (ii) extended HII regions, mixed with non-thermal radiation, likely shaping the structure of the envelope, (iii) clusters of hot molecular cores with a myriad of lines and different chemical and physical properties, and (iv) a converging filamentary structure that transports mass from the outside to the center of the most massive cores. I will present these and some other recent results in the seminar.

Inicio: 12:00h
Ponente: Michele Ronchi (ICE, CSIC)
Lugar: https://us02web.zoom.us/j/85044910172?pwd=bXRmeWx3cTdDSzkrMFFJUXI5eE9hQT09 ; Meeting ID: 850 4491 0172 ; Passcode: 996691

Constraining the birth properties of isolated neutron stars is crucial as it gives information on how these compact objects are formed, how they evolve in time and helps to understand a variety of features we see in the observed population of Galactic neutron stars. In this talk I will present our recent results using machine learning techniques to infer the birth dynamical properties of Galactic neutron stars. I will show that artificial neural networks have the power to estimate with high accuracy the parameters which control the current positions and proper motions of a mock population of pulsars. For this purpose, we have implemented a simplified population-synthesis framework able to simulate populations of pulsars with varying distributions of natal kick velocity and birth distances from the Galactic plane. By evolving the pulsar trajectories in time, we have obtained a series of simulations that are used to train and validate a suitably structured convolutional neural network. I will demonstrate that our network is able to recover from the simulated populations the parameters governing the kick-velocity and Galactic height distribution with a mean relative error of about 0.01. I will discuss the limitations of our idealized approach and present a toy problem to introduce selection effects in a phenomenological way into our simulations by incorporating the information derived from the observed proper motions of 216 isolated pulsars. Our analysis highlights that increasing the sample of pulsars with accurate proper motion measurements by a factor of ~10 we might succeed in constraining the birth spatial and kick-velocity distribution of the neutron stars in the Milky Way with high precision through machine learning.