RSAA Colloquia / Seminars / Feast-of-Facts: Tuesday, 07 December 2021, 11:00-12:00; ZOOM only


Ting-Yun (Sunny) Cheng

"Invited Colloquium: Beyond the Hubble Sequence - What can a machine tell us?"

Nowadays, with current and future astronomical facilities and large-scale sky surveys, hundreds of millions of galaxies are imaged. Conventional techniques for galaxy morphology become challenging, even impossible to apply and analyse numerous datasets generated in the coming era. Hence, machine learning techniques are of paramount importance in future galaxy studies. In this talk, I will start with how supervised machine learning techniques assist in the task of galaxy morphological classification. I will discuss a variety of issues from two perspectives: machine learning and the definition of galaxy morphological types. After this, I will introduce unsupervised machine learning applications to explore galaxy morphology which provides a new insight towards galaxy morphological classification. In particular, machines decide a classification system to describe the variation shown in galaxy morphology in the dataset. The methodology results in 27 machine-defined classes which are physically distinctive from each other. When we merge these machine classes into two main classes for binary classification, the unsupervised method provides an accuracy of 87% for separating early-type (ETG) and late-type galaxies (LTG) using a dataset with the morphology distribution of nearby galaxies (i.e., 22.7% ETG and 77.3% LTG).