Rare Galaxy Classes Identified In Foundation Model Representations
This work addresses the challenge of discovering scientifically-interesting galaxy classes for astronomers, but it is incremental as it applies existing clustering methods to new data from foundation models.
The researchers tackled the problem of identifying rare galaxy populations by analyzing the learned representations of pretrained models, revealing groups with distinctive morphologies beyond the original training objectives.
We identify rare and visually distinctive galaxy populations by searching for structure within the learned representations of pretrained models. We show that these representations arrange galaxies by appearance in patterns beyond those needed to predict the pretraining labels. We design a clustering approach to isolate specific local patterns, revealing groups of galaxies with rare and scientifically-interesting morphologies.