Special Session 78: Special Session on Mathematics of Data Science and Applications
Organizer(s): Ding-Xuan Zhou , Xiang Zhou

Parallel Session 5 :: Tuesday, 12/17, 8:00-10:00                 Capital Suite 21 A
8:00-8:30	Jianbin Yang (Hohai University, China, Peoples Rep of China)  Approximation from Noisy and Blurring Data

8:30-9:00

Yu Cao (Shanghai Jiao Tong University, Peoples Rep of China)  Exploring the Optimal Choice for Generative Processes in Diffusion Models

9:00-9:30

Ting Hu (Xi`an Jiaotong University, Peoples Rep of China)  Pairwise learning problems with regularization networks and Nystrom subsampling approach

9:30-10:00

Zhongjian Wang (Nanyang Technological University, Singapore)  Towards generative diffusion models in infinite dimension

Parallel Session 9 :: Wednesday, 12/18, 8:00-10:00                  Capital Suite 6
8:00-8:30	Yunwen Lei (The University of Hong Kong, Peoples Rep of China)  Stochastic Gradient Methods: Bias, Stability and Generalization

8:30-9:00

Shuyang Ling (NYU Shanghai, Peoples Rep of China)  Beyond Unconstrained Features: Neural Collapse for Shallow Neural Networks with General Data

9:00-9:30

Tao Luo (Shanghai Jiao Tong University, Peoples Rep of China)  The Theory of Parameter Condensation in Neural Networks

9:30-10:00

JIA CAI (Guangdong University of Finance and Economics, Peoples Rep of China)  Enhanced Efficient Heterogeneous Graph Neural Networks

Parallel Session 10 :: Wednesday, 12/18, 12:30-14:30                 Capital Suite 6
12:30-13:00	Yuqing Liu (City University of Hong Kong, Hong Kong)  Error Analysis of Shallow Neural Network on Korobov Space

13:00-13:30

Tong Mao (King Abdullah University of Science and Technology, Saudi Arabia)  Approximation Rates for Shallow ReLUk Neural Networks on Sobolev Spaces via the Radon Transform

13:30-14:00

Nathanael Tepakbong (City University of Hong Kong, Hong Kong)  Solving for the Mean Escape Time with Operator Learning and Deep Neural Networks

14:00-14:30

Ding-Xuan Zhou (dingxuan.zhou@sydney.edu.au, Australia)  The role of structures in neural networks

Parallel Session 11 :: Wednesday, 12/18, 14:45-16:45                  Capital Suite 6
14:45-15:15	Ting Gao (Huazhong University of Science and Technology, Peoples Rep of China)  Critical transitions in brain: modelling and control

15:15-15:45

Huan Lei (Michigan State University, USA)  An energy-stable machine-learning model of non-Newtonian hydrodynamics with molecular fidelity

15:45-16:15

Qianxiao Li (National University of Singapore, Singapore)  Learning, approximation and control

16:15-16:45

Xiang ZHOU (City University of Hong Kong, Hong Kong)  StringNET: Neural Network based Variational Method for Transition Pathways

Parallel Session 12 :: Wednesday, 12/18, 17:00-18:30                Capital Suite 6
17:00-17:30	Chengyu LIU (City University of Hong Kong, Hong Kong)  A Generative Model-Based Variational Method for Wasserstein Gradient Flow