2026 International Conference on Frontiers of Data Science

10:35-11:55 (UTC+8) 10:35-11:55 Local Time | Advances in Machine Learning Theory

Organizer：

Chair：

NO.	Beijing Time (UTC+8)	Local Time	Type	Presentation Topic	Speaker	Affiliation / Organization
1	10:35-10:55	10:35-10:55	Invited Talk	Towards Understanding How Transformers Perform In-Context Logistic Regression	Yuan Cao	The University of Hong Kong
2	10:55-11:15	10:55-11:15	Invited Talk	Non-vacuous Generalization Bounds for Overparameterized Shallow Neural Networks	Yunwen Lei	The University of Hong Kong
3	11:15-11:35	11:15-11:35	Invited Talk	Transformers Meet In-Context Learning: A Universal Approximation Theory	Gen Li	The Chinese University of Hong Kong
4	11:35-11:55	11:35-11:55	Invited Talk	Spectral Gradient Descent Mitigates Anisotropy-Driven Misalignment: A Case Study in Phase Retrieval	Wei Huang	RIKEN AIP and The Institute of Statistical Mathematics

14:00-15:20 (UTC+8) 14:00-15:20 Local Time | Prediction and Machine learning in biomedical studies

Organizer：

Chair：

NO.	Beijing Time (UTC+8)	Local Time	Type	Presentation Topic	Speaker	Affiliation / Organization
1	14:00-14:20	14:00-14:20	Invited Talk	Machine learning methods to predict amyloid positivity using domain scores from cognitive tests	Guogen Shan	University of Florida
2	14:20-14:40	14:20-14:40	Invited Talk	Statistical methods for cell-cell interaction studies on spatially resolved transcriptomics	Xiaoyu Song	Duke-NUS Medical School
3	15:00-15:20	15:00-15:20	Invited Talk	Forecasting Respiratory Infectious Diseases with Adaptive and Physics-Informed Machine Learning	Kam Lun Tsang	The University of Hong Kong
4	15:20-16:40	15:20-16:40	Invited Talk	Joint modeling in presence of informative censoring on the retrospective time scale with application to palliative care research	Quran Wu	Jiangsu Hengrui Pharmaceuticals Co., Ltd.

15:50-17:10 (UTC+8) 15:50-17:10 Local Time | Robust and Efficient Estimation for Modern Causal and Treatment Effect Models

Organizer：

Chair：

NO.	Beijing Time (UTC+8)	Local Time	Type	Presentation Topic	Speaker	Affiliation / Organization
1	15:50-16:15	15:50-16:15	Invited Talk	Rate-Multiply Robust Estimation of General Treatment Models via Balanced Weighting	Wei Huang	University of Melbourne
2	16:15-16:40	16:15-16:40	Invited Talk	Perturbed Double Machine Learning: Nonstandard Inference Beyond the Parametric Length	Mengchu Zheng	Rutgers University
3	16:40-17:05	16:40-17:05	Invited Talk	Towards an efficiency theory on parameter manifolds	Lin Liu	Shanghai Jiao Tong University

2026-05-16 (Saturday)