Photometric Data-driven Classification of Type Ia Supernovae in the Open Supernova Catalog
This addresses the challenge of accurate supernova classification for astronomers, but is incremental as it builds on existing machine-learning approaches with a specific data focus.
The researchers tackled the problem of classifying Type Ia supernovae using only real photometric observation data for training, achieving good results on real data from the Open Supernova Catalog despite a small training sample. They also found that transferring models between simulated and real data significantly reduces performance, highlighting differences between these data types.
We propose a novel approach for a machine-learning-based detection of the type Ia supernovae using photometric information. Unlike other approaches, only real observation data is used during training. Despite being trained on a relatively small sample, the method shows good results on real data from the Open Supernovae Catalog. We also investigate model transfer from the PLAsTiCC simulations train dataset to real data application, and the reverse, and find the performance significantly decreases in both cases, highlighting the existing differences between simulated and real data.