A Deep Learning Approach for Characterizing Major Galaxy Mergers
This work addresses a key problem in astronomy for validating galaxy formation theories, but it is incremental as it applies an existing deep learning method to a new domain with specific gains.
The paper tackles the problem of estimating galaxy merger stages from single images by introducing a CNN-based regression model that predicts the merger stage relative to the first perigee passage with a median error of 38.3 million years over a 400 million-year period, and shows reasonable performance on real observations.
Fine-grained estimation of galaxy merger stages from observations is a key problem useful for validation of our current theoretical understanding of galaxy formation. To this end, we demonstrate a CNN-based regression model that is able to predict, for the first time, using a single image, the merger stage relative to the first perigee passage with a median error of 38.3 million years (Myrs) over a period of 400 Myrs. This model uses no specific dynamical modeling and learns only from simulated merger events. We show that our model provides reasonable estimates on real observations, approximately matching prior estimates provided by detailed dynamical modeling. We provide a preliminary interpretability analysis of our models, and demonstrate first steps toward calibrated uncertainty estimation.