Sui Tang

Assistant Professor
Department of Mathematics
University of California, Santa Barbara
552 University Rd, Isla Vista, CA 93117
Email:suitang@ucsb.edu

I am a tenure-track assistant professor at Department of Mathematics. Before joining UCSB, I was a member of Prof. Mauro Maggioni's research group and worked as assistant research professor at Johns Hopkins University. Here is my CV. I am interested in solving problems in the mathematical foundations of Data Science. My current research projects are motivated by the need to exploit dynamical data sets in complex physical systems to perform inference with provable performance and build generalizable and interpretable predictive models, where I used and developed techniques from Statistical/Machine Learning, Harmonic Analysis, Approximation Theory, and Probability. My research can be categorized to the following two areas:

Mathematical foundation of learning theory: statistical learning theory, statistical inference for ODEs,SDEs and PDEs from time-series data, high dimensional data analysis
Applied and computational harmonic Analysis: functional analysis, Fourier analysis, approximation theory, sampling and frame theory, inverse problem in mathematical/statistical signal processing

I am spending part of Fall 2021 as a visiting scientist at Simons Institute for the Theory of computing, and will be participating in the program Geometric Methods in Optimization and Sampling". Here is my talk at Simons intsitute workshop about my recent work in stocastic systems of interacting particles.

Current teaching

Math 260H: Seminar in mathematical foundation of data science
Math 104A: Numerical Analysis

Graduate and Undergraduate Research Opportunities

I am looking for highly self-motivated graduate students and senior undergraduate students to work on projects generally related to machine learning, dynamical system, signal processing, and data analysis. Please email me your CV (including projects you have done), transcript (if applicable) if you are interested.

Grants and Awards

Hellman Faculty Fellows 2022-2023
NSF DMS-2111303, 2021-2024
Simons Collobration Grant for Mathematitians, declined due to the NSF award
Early Career Faculty Acceleration Grant, 2021
Regents Junior Faculty Fellowship, 2021
Best Overall Award in the poster competition of Second International Conference on Mathematics of Data Science, 2018
AMS-Simons Travel grant, 2017-2020

Recent travels/talks

Conference on Advances in data science: theory, methods and computation. TAMU, Oct 2022
Kinetic Equations: Recent Developments and Novel Applications, Banff workshop, Oaxaca, Nov 2022
UC Berkeley Applied math seminar, Nov 2022
Special session on Mathematics of information theory, 2022 Pacific Rim Mathematical Association Congress, Vancouver, Dec 2022
AMS Special Session on Current Progress in Computational Biomedicine, joint meeting, Boston, Jan 2023
special session at 13th AIMS Conference on Dynamical Systems, Differential Equations and Applications, Wilmington, NC USA May, 2023.
"Topics on Neuroscience, Collective Migration and Parameter Estimation", Mathematical Institute at the University of Oxford, July 2023.
International Council for Industrial and Applied Mathematics (ICIAM), Tokyo, Japan, Aug 2023
Isaac Newton Institute workshop "Measures and Representations of Interactions", Cambridge, UK, Sep 2023.

Publications: Google Scholar

Learning Transition Operators From Sparse Space-Time Samples (with Christian Kummerel and Mauro Maggioni), 1-35,submitted. Arxiv 2212.00746
A Hybrid Predictive Model of COVID-19 Cases using Autoregressive Model and LSTM (with Yangyi Zhang and Guo Yu), 1-19,submitted. Arxiv:2211.17014
Higher-order error estimates for physics informed neural networks approximating primitive equations (with Ruimeng Hu, Quyuan Lin and Alan Rayden). 1-27, submitted. Arxiv:2209.11929
Robust recovery of bandlimited graph signals via randomized dynamical sampling (with Longxiu Huang and Deanna Needell). ArXiv:2109.14079,1-33, Submitted.
Learning particle swarm models from data with Gaussian process (with Jinchao Feng, Charles Kulick, and Yunxiang Ren),1-44, Submitted. A short version is accepted as a poster presentation at Neurips Workshop 2021 Machine Learning and the Physical Sciences.
Learning theory for inferring interaction kernels in second-order interacting agent Systems (with J. Miller, M. Zhong and M. Maggioni). ArXiv:2010.03729, 1-68, Submitted.
Scalable marginalization of latent variables for correlated data with applications to learning particle interaction kernels, (with Mengyang Gu, Xubo Liu, Xinyi Fang), 1-14, to appear at The New England Journal of Statistics in Data Science, 2022
A Numerical Study on Sparse Learning of Interaction Laws in Homogeneous Multiparticle Systems with (Hao-Tien Chuang, Dongyang Li, Shelby Malowney, and Ritwik Trehan), SIAM Undergraduate Research Online, 232-249,2022. Code.

Estimate the spectrum of affine dynamical systems from partial observations of a single trajectory data (with Jiahui Cheng, undergraduate student). To appear at Inverse Problems. Code

Learning interaction kernels in stochastic systems of interacting particles from multiple trajectories (with F. Lu and M. Maggioni). To appear at Foundations of Computational Mathematics, 1-55, 2021.
Learning interaction kernels in heterogeneous systems of agents from multiple trajectories (with F. Lu and M. Maggioni). Journal of Machine Learning Research, 22 (32), 1-67, 2021. Code.

On the identifiability of interaction functions in systems of interacting particles (with Z. Li, F. Lu, M. Maggioni and C. Zhang). Stochastic processes and their applications, Vol 132, 135 - 163, 2021.
Sensor Calibration for Spectral Estimation Off the Grid (with Y. Eldar and W. Liao). Applied and Computational Harmonic Analysis,48(2), 570-598, 2020.
Phaseless reconstruction from space-time samples (with A. Aldroubi and I. Krishtal). Applied and Computational Harmonic Analysis, 48(1), 395-414, 2020.
Nonparametric inference of interaction laws in systems of agents from trajectory data (with F. Lu, M. Zhong and M. Maggioni). Proceedings of the National Academy of Sciences of USA, 116(29): 14424-14433, 2019. Appendix: 27 pages.
Analysis of simulated crowd flow exit data: visualization, panic detection, exit time convergence, attribution and estimation (with A. Grim, B. Iskra, N. Ju, A. Kryshchenko, F.P. Medina, L. Ness, M. Ngamini, M. Owen, R. Paffenroth), Research in Data Science, 17:239-281, 2019.
Undersampled windowed exponentials, spectra of Toeplitz operators and its applications (with C. Lai). Acta Applicanda Mathematicae, 164(1), 65-81, 2019
Universal spatial-temporal sampling sets for discrete spatially invariant evolution systems. IEEE Transactions on Information Theory, 63(9): 5518-5528, 2017.
System Identification in Dynamical Sampling. Advance in Computational Mathematics, 43(3): 555-580, 2017.
Dynamical Sampling (with A. Aldroubi, C. Cabrelli and U. Molter). Applied and Computational Harmonic Analysis, 42(3): 378-401, 2017.
Phase retrieval of evolving signals from space-time samples (with A. Aldroubi and I. Krishtal). Proceeding of 12th international conference on Sampling Theory and Applications, 2017.
Multidimensional Signal Recovery in Discrete Evolution Systems via Spatiotemporal Trade Off (with R. Aceska and A. Petrosyan). Sampling Theory in Signal and Image Processing, 14(2):153-169, 2015.
Filter Recovery in Infinite Spatially Invariant Evolutionary Systems via Spatiotemporal Trade off. Proceeding of 11th international conference on Sampling Theory and Applications, 2015.
Dynamical sampling of two-dimensional temporally-varying signals (with R. Aceska and A. Petrosyan). Proceeding of 11th international conference on Sampling Theory and Applications, 2015.
Dynamical sampling in hybrid shift invariant spaces (with R. Aceska). Contemporary Mathematics, American Mathematics Society, Providence, 626:149-166, 2014.