Orateur
Description
The original Photometric LSST Astronomical Time-Series Classification Challenge (PLAsTiCC) established the gold-standard reference dataset of Rubin light curves and catalyzed the development of several deep learning algorithms for classification. However, to truly simulate LSST operations, we must simulate not light curves, but real-time alert streams, complete with the contextual information an astrophysicist would receive from Rubin. This dataset will allow the Rubin ML community to prepare for the start of commissioning, as well as provide a new reference sample for cosmological investigations with LSST. The key goal of ELAsTiCC to prepare LSST's Science Collaborations working in the time-domain for Rubin Operations. In creating this successor to PLAsTiCC, we have incorporated deep-learning in generating the simulations, and we expect the simulations to become the basis for many novel ML methods that encode both light curves and contextual information into a complex feature space. I will discuss the ELAsTiCC team's work in building this dataset, and progress towards launching the challenge, and discuss several of the studies that we have envisioned for this dataset.
