Monica Valluri

The focus of the research in our group is two fold. 1) We use stellar kinematics in galactic nuclei to infer the masses of super massive black holes using dynamical modeling. This includes a combination of statistical and optimization methods to fit data from telescopes such as the James Webb Space telescope. My group also uses galactic dynamical simulations to study the effects that black holes have on their host galaxies and statistical methods to compare observed galaxies with simulations. 2) We construct dynamical models of the Milky Way to understand its structure, dynamics and formation history. We construct models with 9-dimensional data (3D positions, 3D velocities, iron-abundance, alpha-element abundance, age) for tens of millions of individual stars in the Milky Way to constrain the properties of dark matter. Dark matter is the substance that constitute 85-90% of the mass budget in the Universe but is still undetected. The data science methods used here including unsupervised learning to identify clusters in 6D-9D phase space – which enables us to identify the satellites that built up the Milky Way and the nature of the dark matter in these satellites, Bayesian inference to constrain the nature of dark matter by characterizing the structure of tidal streams, deep learning to infer properties like ages and distances of stars from their spectra.