Pushing Rare Event Search to the New Limit with Model- and Data-Centric AI

by Prof. Aobo Li (UC San Diego)

Friday 12 Jul 2024, 10:00 → 12:00 Asia/Shanghai

Abstract:

Rare event searches allow us to search for new physics inaccessible with other means by leveraging specialized radiation detectors. Artificial Intelligence provides a new tool to maximize the information provided by these detectors. The information is sparse, which forces us to start from the lowest level data and design customized models to produce results. This seminar will focus on two main areas within rare event search experiments: neutrinoless double beta decay and axion dark matter search. For the former, we will delve into the KamLAND-Zen experiment with its monolithic liquid scintillator detector, discussing our work on model-centric AI, which focuses on designing optimal algorithms to significantly boost KamLAND-Zen’s sensitivity. For the latter, we will present our recent work on data-centric AI, highlighting the release of data from the ABRACADABRA experiment, which enables core AI algorithms to directly produce real physics results. In the latter part of the presentation, we will explore the potential of next-generation AI and data science tools that are being developed to fully realize the discovery capabilities of rare event search experiments.

 

About the Speaker:
Prof. Aobo Li is an assistant professor at UC San Diego Halicioglu Data Science Institute and Department of Physics. He received his PhD in Physics from Boston University. His work lies at the intersection between machine learning and experimental particle and nuclear physics, especially rare event search experiments. His work has received several awards including the APS Dissertation Awards in Nuclear Physics.

----------------------------

Meeting ID 会议号: 85078311284
Meeting URL 会议链接:: https://us02web.zoom.us/j/85078311284?pwd=GlmFYmxdRDmqe5JwgjB40nOdOrffiF.1
HostKey 主持人密钥: 068014
Password 会议密码: 240712