On May 20, 2025, at the invitation of the School of Artificial Intelligence, Shenzhen University, Assistant Professor Pang Guansong from Singapore Management University (SMU) delivered a lecture entitled "Generalist Models for Anomaly Detection." The lecture was hosted by Professor Zhu Zexuan, Deputy Director of the National Engineering Laboratory for Big Data, and attended by approximately 20 faculty and students.
Professor Pang introduced the latest methods and empirical results for generalist anomaly detection in zero-shot and few-shot settings. The core idea is to construct a generalist anomaly detection model, i.e., training a single detection model that can be generalized to detect anomalies in various datasets from different application domains without any further training/adjustment to the target data. A lively discussion followed.
This lecture not only provided the faculty and students with cutting-edge research ideas but also inspired reflection on future research topics.
Guest Introduction:
Dr. Pang Guansong is a tenured Assistant Professor of Computer Science and a Lee Kong Chian Research Fellow at the School of Computing and Information Systems, Singapore Management University (SMU), leading the Machine Learning and Applications (MaLA) Lab. He is also a faculty member at the Centre for Security, Mobile Applications, and Cryptography. He was a researcher at the Australian Institute for Machine Learning (AIML) at the University of Adelaide, Australia. Prior to joining AIML, he received his PhD from the University of Technology Sydney (UTS). His research interests include machine learning, data mining, and computer vision, with a focus on identifying and generalizing anomalous/unknown/unseen data to create trustworthy artificial intelligence systems. His research has been cited over 9,600 times and has received numerous global recognitions and awards, such as inclusion in the prestigious 2020 UTS Vice-Chancellor's Award list, being named one of the top 2% of scientists globally for 2022-2024, receiving the DSAA 2023 Best Paper Award, and the KDD 2023 Most Influential Paper Award. He is actively involved in various professional activities, serving as an area chair for NeurIPS, ICLR, ICML, CVPR, KDD, PAKDD, and IJCAI; an associate editor for the IEEE Transactions on Neural Networks and Learning Systems (TNNLS) and the Journal of Pattern Recognition; and an editorial board member for IEEE Intelligent Systems and the International Journal of Data Science and Analytics.
