| |
 | | Date: | 7 November 2025, Friday | | Time: | 3:45pm to 4:45pm | | Venue: | School of Computing & Information Systems 1 (SCIS1), Level 2, Seminar Room 2-4, Singapore Management University, 80 Stamford Road, Singapore 178902 |
|
Limited seating. Registration will close on 26 October 2025 or once maximum capacity is reached. Registration is required for attendance. Light refreshment will be provided after the talks. | Research Cluster: Artificial Intelligence & Data Science | | | | Topic: | Selecting Comparative Sets of Reviews Across Multiple Items | | Speaker: | Hady W. LAUW, Associate Professor of Computer Science | | Abstract: | While choosing among several products, users may look up reviews from each product they are considering. Due to the large number of reviews of products, selecting representative reviews from one product alone is already a challenging problem. In this work, we further aim to conduct review selection for multiple products simultaneously for comparative purposes. We formulate objective functions that synchronize the review selection and design efficient algorithms to optimize for the objective functions. To narrow down the potentially long list of comparison items into a shorter list of more similar items, we construct a graph representing items’ similarity and design efficient algorithms to find the heaviest 𝑘-subgraph including the target item. The results are validated on real world datasets on various product categories. |
| | | | Topic: | A New Framework for Personalized Federated Learning | | Speaker: | Zhize LI, Assistant Professor of Computer Science | | Abstract: | Federated Learning (FL) enables multiple clients to collaboratively train models without sharing raw data, but the non-IID data across clients often limits the performance of a single global model. Personalized Federated Learning (PFL) has emerged as a promising approach to address this challenge, typically by combining a shared representation with a personalized head. However, its effectiveness still critically depends on the quality of the shared representation. In this talk, I will introduce FedROS, a new PFL framework that enhances the shared representation through a personalized orthogonal sparse complement for each client. FedROS achieves faster linear convergence for both shared and personalized components while requiring fewer data samples per client. Experiments on real-world datasets demonstrate that FedROS consistently outperforms existing PFL methods, validating its effectiveness and scalability. |
| | | | Topic: | Goal driven Interactive AI to Improve Human Ability | | Speaker: | Pradeep Reddy VARAKANTHAM, Professor of Computer Science | | Abstract: | In this talk, I will first give a broad overview of our research conducted on the broad theme of goal driven interactive AI as part of the CARE.AI lab. Then, I will delve deeper into one topic on Safe Value Alignment, a key area of interest within this broad theme of goal driven interactive AI. |
| | | | ABOUT THE SPEAKER(S) | | |  | Hady W. Lauw is an Associate Professor as well as the Director of BSc (Computer Science) Programme at the School of Computing and Information System of Singapore Management University (SMU). His research group Preferred.AI is active in artificial intelligence, data mining, and machine learning, focusing on Web mining, preference analytics and recommender systems. He has been awarded National Research Foundation Fellowship, Lee Kong Chian Fellowship, and Educational Research Fellowship. He is active in the data mining and Web communities, recently serving as Program Chair of KDD-25 and WWW-24 as well as General Chair of WSDM 2023. | | |  | Zhize Li is an Assistant Professor of Computer Science at Singapore Management University. Prior to that, he was a Research Scientist at Carnegie Mellon University. He received his PhD degree from the Institute for Interdisciplinary Information Sciences at Tsinghua University in 2019, where his PhD thesis won the Tsinghua Outstanding Doctoral Dissertation Award. He was named a Rising Star in AI by KAUST in 2022. His research mainly focuses on optimization, federated learning, AI privacy, and machine learning. | | |  | Pradeep Varakantham is a Professor of Computer Science at School of Computing and Information Systems at Singapore Management University (SMU). He directs the Collabrative and Trustworthy Artificial Intelligence (CARE.AI) lab at SMU that is focussed on developing AI systems that improve human decision making in a wide variety of domains including education, scam prevention, public health and others. He has published more than 150 research papers at top conferences and journals in Artificial Intelligence and has also won recognitions/awards at top conferences in AI. He is a AAAI senior member, won Google grant for social impact and was a Lee Kuan Yew fellow from 2022-2025. He also coordinates the BSc AI track at SMU teaches AI courses for undergraduates, masters and PhD students at SMU. | | | | SEMINAR MODERATOR | | |  | LIM Ee-Peng
Professor of Computer Science
Director, Artificial Intelligence & Data Science Cluster |
|
|
|
