aphearin/chopped_data_distribution

Disposable development code for distributing simulated data with mpi4py and thechopper

About me

I am a Physicist in the Cosmological Physics & Advanced Computing (CPAC) Group at Argonne National Laboratory. I study cosmology and astrophysics using numerical simulations, theoretical models of cosmological structure growth, machine learning techniques, and observations from large-area galaxy surveys.

I am the lead developer of Halotools, an open-source python package for creating model universes of galaxies. I am also a member of several large collaborative experiments designed to measure dark energy and dark matter, including the Dark Energy Spectroscopic Instrument (DESI), and the Vera Rubin Observatory Legacy Survey of Space and Time Dark Energy Science Collaboration (LSST DESC).

Research

My recent work focuses on developing simulation-based, differentiable forward models of cosmological structure growth in the deeply nonlinear regime. The goals of this effort are to help uncover the physical nature of dark matter and dark energy, and to understand the connection between galaxies and dark matter halos. This program involves many different components, all of which leverage the autodiff library JAX, including:

• Diffmah, a model of the assembly of individual dark matter halos
• Diffstar, a model of the star formation history of individual galaxies
• Diffprof, a model of the evolution of the density profile of dark matter halos
• DSPS, a differentiable library of stellar population synthesis

Contact information

Dr. Andrew Hearin

email: ahearin [at] anl [dot] gov