- Nov. 1, 2021, 2:00 pm US/Central
- Robert Morgan, University of Wisconsin
Abstract:The current tension on the measured value of the expansion rate of the Universe today (H0) has pushed the cosmological community to pursue more measurement techniques and larger datasets. One such technique, the characterization of systems with gravitationally lensed supernovae, is of particular interest in the Rubin Observatory era for two reasons: (1) thousands of these systems will be detectable by the Rubin Observatory when only a handful exist in current datasets, and (2) each identified system offers an independent measurement of H0, enabling tight constraints over the 10-year Legacy Survey of Space and Time. In this talk, I present a novel deep learning architecture for lensed supernovae detection that identifies more lensed supernovae (90 percent of the expected population) than any other previously proposed method.