Title：Astronomy in the big data era: exploring galaxy morphology with citizen science and machine learning - a case study

Speaker：Nan Li (NAOC)

Time：3:00 pm March 28th (Thursday)

Tencent Meeting：42915400486 password: 6360

Location: Lecture Hall, 3rd floor

Report in English

Abstract：

The exponential growth of astronomical datasets provides an unprecedented opportunity for humans to gain insight into the Universe. However, effectively analyzing this vast amount of data poses a significant challenge. As a case study, I will present a series of investigations about exploring galaxy morphology from enormous astronomical datasets in this talk. Citizen science, such as the Galaxy Zoo and Spaceways projects, is one way to tackle this problem. To make it easier for citizens to participate, we have developed three new citizen science projects with more advanced user interfaces, such as Web-UI and smartphone apps. Besides, I will also discuss studying galaxy morphogies with machine learning algorithms, including supervised, unsupervised, semi-supervised, and self-supervised learning. At last, I may introduce an AI framework to achieve efficient Human-machine cooperation by combining a Large Vision Model and the Human-in-the-loop mechanism, which exhibits notable few-shot learning capabilities and versatile adaptability to astronomical vision tasks beyond galaxy morphology classification, which can be the "hammer" to deal with the "nails" mentioned at the beginning.

CV：Dr. Nan Li is a faculty researcher (2020-present) at the National Astronomical Observatories, Chinese Academy of Sciences (NAOC). He obtained his Ph.D. at NAOC in 2013, then conducted his postdoctoral search at the University of Chicago  (2013-2017) and the University of Nottingham (2017-2019). His research focuses on dark matter, galaxy evolution, and AI4Astronomy.