About Me
Hi, I'm Xinyue! Welcome to my personal website.
I'm currently a postdoctoral researcher (and previously a PhD student) at the Astrophysics Research Centre, Queen's University, Belfast, Northern Ireland, United Kingdom.
My research interests include machine learning, Rubin/LSST, supernovae, and nuclear transients.
I grew up in CHANGCHUN city in northeastern China -- the capital of JILIN province, where the people are known for their hospitality and love for food!
*Fun fact: CHANGCHUN means "long spring" in Chinese, but it is actually a city with very long winter! 🙈
Research
Full publications available on my NASA/ADS library
NEEDLE: Neural Engine for Discovering Luminous Events
I am the lead developer of a multi-modal machine-learning pipeline designed for LSST-era alert streams to identify rare transients such as Superluminous Supernovae and Tidal Disruption Events from real-time survey data. Predictions are publicly distributed through the Lasair UK alert broker and have contributed to 29 confirmed discoveries since Aug.~2024.
[Paper I]
[Paper II]
SRNN Framework for LSST AGN Time-Series Analysis
Implementation of a Stochastic Recurrent Neural Network (SRNN) for modelling simulated AGN light curves under LSST Wide-Fast-Deep cadences using CARMA variability models, demonstrating recovery of variability structure functions while highlighting cadence limitations.
Contact
Email: x.sheng@qub.ac.uk
Address: School of Mathematics and Physics, Queen's University Belfast. BT7 1NN, Belfast, Northern Ireland, United Kingdom
