报告题目：Case Studies of AI+DB Technologies: Learned Indexes and Large Language Models
主讲人： 王炜 香港科技大学（广州）教授
报告时间：2023年08月17日（星期四）10:00-11:00
报告地点：海韵园行政楼A306
报告摘要：
The rapid development of AI technologies expands its impact into multiple fields, including the data management domain. In this talk, we will first present our recent work in integrating AI and DB technologies on the emerging topic of learned indexes. While learned indexes for one-dimensional data have been widely studied, there are relatively fewer results on learned indexes for multi-dimensional and high-dimensional data. In this talk, we will introduce new learned indexes for these two types of data, respectively. For the multi-dimensional data, we propose an index based on a learned space-filling curve, together with novel query optimization algorithms. For the high-dimensional data, we discuss two types of integrating learning into indexing and approximate k-NN query processing, including one that is optimized for external I/Os and the other that predicts the parameters used for probing the index. Next, we will briefly introduce the Large Language Models (LLMs) and some of its impact to the data management field.
报告人简介：
Dr. Wei Wang is a currently a Professor in the Data Science and Analytics Thrust, Information Hub, The Hong Kong University of Science and Technology (Guangzhou), China. Before that, he was a Professor in the School of Computer Science and Engineering, The University of New South Wales, Australia. His current research interests include Similarity Query Processing, Artificial Intelligence, Knowledge Graphs, Security for AI Models, and AI for Science.
He has published more than 180 papers in reputed journals and conferences, and has won the Best Paper Awards in ICASSP 2023, SIGCOMM 2022, ICMR 2021, and the Best Student Paper at DASFAA 2016. He is an Associate Editor of IEEE Transactions on Knowledge and Data Engineering and Journal of Materials Informatics, and program committee members in various first-tier conferences (SIGMOD, VLDB, ICDE, SIGIR, SIGKDD, etc.).
邀请人：软件工程系 王晓黎副教授