Time：Monday, April 21st, 2025, 4:00 PM
Place： X10931
Introduction：
The rapid expansion of sensor deployments across industries such as manufacturing, aerospace, and healthcare has created a growing need for AI-driven time series analytics. This talk explores recent advancements in AI techniques that enable predictive maintenance, optimize machine health monitoring, and enhance operational efficiency. We discuss key challenges and solutions in three major areas: (1) self-supervised representation learning for time series data, which improves feature extraction from unlabeled data using contrastive learning; (2) unsupervised domain adaptation for multivariate sensor data, addressing distribution discrepancies at both local and global levels to enhance cross-domain generalization; and (3) model compression techniques for efficient edge deployment, ensuring AI models can operate in resource-constrained environments. Additionally, we examine how foundation models for time series analytics can be adapted to downstream applications.
Delivered by Xiaoli Li
Xiaoli is currently the Department Head and Senior Principal Scientist at the Institute for Infocomm Research, A*STAR, Singapore. He also serves as an adjunct full professor at the School of Computer Science and Engineering, Nanyang Technological University, Singapore. With a diverse range of research interests, Xiaoli focuses on cutting-edge areas such as AI, data mining, machine learning, and bioinformatics. His contributions to these fields are evident through his extensive publication record, boasting over 360 peer-reviewed papers, and the recognition he has received, including over ten best paper awards. He has been serving as Editor-in-chief of the Annual Review of Artificial Intelligence and an Associate Editor for prestigious journals like IEEE Transactions on Artificial Intelligence and Knowledge and Information Systems, as well as conference chairs and area chairs of leading AI, machine learning, and data science conferences, such as AAAI, IJCAI, ICLR, NeurIPS, KDD, ICDM etc. Beyond academia, Xiaoli possesses extensive industry experience, where he has successfully spearheaded over 10 R&D projects in collaboration with major industry players across diverse sectors, such as aerospace, telecom, insurance, and professional service companies. Xiaoli is an IEEE Fellow and Fellow of Asia-Pacific Artificial Intelligence Association (AAIA). He has been recognized as one of the world's top 2% scientists in the AI domain by Stanford University, and Clarivate's Highly Cited Researcher.
