Fast spectral tools for graph structure.
Generalizing classic eigenvalue localization results (e.g. Gershgorin).
Theory and scalable algorithms for kernel-based function approximation.
Asynchronous parallel algorithms for finding minima fast by fitting functions to surrogate models.
2019-05-20: Teaching Numerical Methods for Data Science at SJTU
Numerical Methods for Data Science (SJTU CS 259)
Short course offered May-June 2019 at SJTU.
Survey course of numerical methods prominent in modern data analysis and machine learning. Building on basic methods of optimization and numerical linear algebra, the course will explore the role of numerical methods for treating several classes of data analysis problems, including low rank factorizations and completion of matrix data; function approximation by kernel methods; and analysis of data on graphs.
SCAN Seminar (CS/MATH 7290)
Ongoing. M 1:25-2:15, Gates 406.
The Scientific Computing and Numerics seminar series focuses on various methods in scientific computing, the analysis of convergence properties and computational efficiency, and their adaptation to specific applications.