Event Details

Galaxy Zoo in the Deep Learning Era by Mr Mike Walmsley

Abstract:

Deep learning is fundamental to Galaxy Zoo’s latest morphology catalogs. In this talk, I explore our new capabilities to accurately predict how GZ volunteers would describe a galaxy, and consider what these capabilities mean for observational astronomers. One clear consequence is scale; we will shortly release detailed morphological classifications for all 1.4 million nearby galaxies in the DESI Legacy Surveys. The statistical power from this scale will unpick the complex relationships between morphology, star formation, and environment. Further, a fundamentally new opportunity is answering your own morphology questions with our public code and models. Having already learned to answer every GZ question at once, our models are easy to adapt to new surveys and new questions. This was recently exploited to create new (and order-of-magnitude larger) catalogs of mergers in HST and HSC, and of ringed galaxies in DESI LS. Finally, we describe our very latest work on simultaneously learning from labelled and unlabelled galaxy images. Such approaches are ideally suited to Euclid and Rubin because they allow us to leverage both the millions of human labels collected over the last decade and the raw scale of unlabelled images these new surveys will produce.