AIMS Home

Special Session on Mathematics of Data Science and Applications

Organizer(s):
Name:	Affiliation:	Country:
Ding-Xuan Zhou	dingxuan.zhou@sydney.edu.au	Australia
Xiang Zhou	City University of Hong Kong	Hong Kong

Introduction:

Over the last twenty years, data science, machine learning, and deep learning in particular, have begun transforming the global economy and modern life. While much attention is focused on empirical data mining success, there have been considerable mathematical structures and a growing body of mathematical theories about how the structures relate to observable properties of real-world systems. Discovering such structures may lead to important mathematical insights and implications for practitioners. This special session aims at interactions among approximation theory, deep neural networks, harmonic analysis, machine learning, numerical analysis, and statistics to foster further research in the fast-developing area of data science.

All Abstracts

List of abstracts and speakers
Name	Affiliation	Title
JIA CAI	Guangdong University of Finance and Economics	Enhanced Efficient Heterogeneous Graph Neural Networks
Yu Cao	Shanghai Jiao Tong University	Exploring the Optimal Choice for Generative Processes in Diffusion Models
Ting Gao	Huazhong University of Science and Technology	Critical transitions in brain: modelling and control
Ting Hu	Xi`an Jiaotong University	Pairwise learning problems with regularization networks and Nystrom subsampling approach
Huan Lei	Michigan State University	An energy-stable machine-learning model of non-Newtonian hydrodynamics with molecular fidelity
Yunwen Lei	The University of Hong Kong	Stochastic Gradient Methods: Bias, Stability and Generalization
Qianxiao Li	National University of Singapore	Learning, approximation and control
Shuyang Ling	NYU Shanghai	Beyond Unconstrained Features: Neural Collapse for Shallow Neural Networks with General Data
Chengyu LIU	City University of Hong Kong	A Generative Model-Based Variational Method for Wasserstein Gradient Flow
Yuqing Liu	City University of Hong Kong	Error Analysis of Shallow Neural Network on Korobov Space
Tao Luo	Shanghai Jiao Tong University	The Theory of Parameter Condensation in Neural Networks
Tong Mao	King Abdullah University of Science and Technology	Approximation Rates for Shallow ReLUk Neural Networks on Sobolev Spaces via the Radon Transform
Nathanael Tepakbong	City University of Hong Kong	Solving for the Mean Escape Time with Operator Learning and Deep Neural Networks
Zhongjian Wang	Nanyang Technological University	Towards generative diffusion models in infinite dimension
Jianbin Yang	Hohai University, China	Approximation from Noisy and Blurring Data
Ding-Xuan Zhou	dingxuan.zhou@sydney.edu.au	The role of structures in neural networks
Xiang ZHOU	City University of Hong Kong	StringNET: Neural Network based Variational Method for Transition Pathways