Anshuman is a computational astrophysicist focussed on studying the interplay between cosmological models and astrophysical processes, utilising novel statistical techniques (Simulation Based Inference, Machine Learning, etc.) and cosmological modelling.
He completed his PhD at the Max Planck Institute for Astrophysics, Garching, Germany, under the supervision of Benedetta Ciardi and Volker Springel, where he worked with the LOFAR telescope's Epoch of Reionisation Science Working Group. As a part of the group, he worked on multiple projects on improving the theoretical understanding of the evolution of properties of the Inter-Galactic Medium (IGM) during the formation of the first galaxies. He developed, improved and ran cosmological simulations of the IGM building the Polar simulations suite, and used simulated outputs from them to build robust Machine Learning based signal extraction approaches. Prior to this, he did his Bachelor's and Master's at IISER Mohali, India, majoring in Physics with a minor in Astronomy.
At Berkeley, he is a BCCP Fellow working with Simone Ferraro, and a part of the Learning the Universe collaboration. He is involved in implementing Simulation Based Inference approaches to cross-correlate CMB observations with galaxy surveys, and use them to derive the best-fit models of the Universe. He is also involved in exploring super-resolution techniques for cosmological simulations, using traditional statistical techniques as well as AI-based methods.
Publications
A complete list of publications can be found here: https://ui.adsabs.harvard.edu/public-libraries/JsvARhIuTLSTp57XeVjEuA
Epoch of Reionization, Early Universe Cosmology, Machine Learning