Deep Learning for Analyzing Big Data from Telescopes

A unified deep learning framework for low-latency analysis of the raw big data collected by our observational instruments can enable real-time multimessenger astrophysics. We plan to train real-time classifiers, based on deep neural networks, with simulations injected into raw image data to search for kilonova from neutron star mergers and supernovae, using the Dark Energy Camera (DECam), Fermi, and possibly the James Webb Space Telescope (JWST) and the Large Synoptic Survey Telescope (LSST).

Project Members: Daniel George, Eliu Huerta