Résumé / Abstract Seminaire_IAP
« Machine learning and exoplanet demographics with the TESS mission »

David Armstrong

The exoplanet field has grown rapidly with the advent of space-based observatories, moving from the discovery of single planetary systems to an era of demographic studies and large surveys. In this talk I will present the results of a long-term program to identify the demographics of exoplanets revealed by the NASA TESS mission. To do this, we developed a machine learning planet vetting and validation pipeline, RAVEN. I will discuss the challenges and opportunities of machine learning in planet validation, and the results of our pipeline on over 2 million TESS lightcurves, discovering over 1000 new vetted candidates and validating over 100 new planets. The combination of a homogeneous pipeline and the large TESS dataset reveals the occurrence rate of planets at close orbital periods (<16d) with a magnitude smaller uncertainties compared to Kepler results. I will conclude by presenting the occurrence rate of planets in and around the Neptunian desert, allowing a deeper understanding of how these planets formed and reached their present state.
vendredi 30 janvier 2026 - 11:00
Amphithéâtre Henri Mineur, Institut d'Astrophysique de Paris
Page web du séminaire / Seminar's webpage