Gamma-Ray Burst Classification through Machine Learning

Gamma-Ray Bursts (GRBs) are among the most energetic events in the universe. Prompt, multiwavelength, and multimessenger observations of these enigmatic transients allow us to probe physics beyond extremes that can be achieved in terrestrial laboratories. Much work remains to fully elucidate the origins of GRBs and to understand their physical processes. A holy grail in GRB studies is prompt classification of GRBs, aiding follow-up by guiding both observing profiles, prioritization of telescope time, and providing key statistical information for follow-up analysis.

We propose to build a fast GRB classification tool, based on modern unsupervised deep learning techniques and relying on a set of inputs for each event, which contains the core information relevant for GRB classification (including duration, temporal variation, pulse structure, spectral hardness and evolution, and how these parameters relate), and have not yet been explored within the machine learning framework: the GRB waterfalls.