Speaker
Description
Astrophysical research is increasingly leveraging data science, statistics and machine learning, with applications ranging from detecting transients in images to inferring exoplanet atmospheres. However, studying the most energetic phenomena in the universe presents particular challenges, including comparatively small, heterogeneous datasets and a lack of reliable training data. This talk provides a brief overview of the work my group is doing in statistics and machine learning for astronomy. I will highlight recent opportunities and challenges in integrating machine learning into astrophysical data analysis, as well as connections to other areas of astronomy. I will show recent results and ongoing work exploring how machine learning can uncover the origin of the most extreme astrophysical phenomena, such as accreting black holes, neutron stars, and the mysterious sources known as Fast Radio Bursts.
