I am a Professor of Statistics at the University of California, Riverside. My research spans high dimensional statistics, machine learning, algorithms and privacy.
I am on Sabbatical leave from July 2024 to Sept. 2025, except for the Winter quarter of 2025, during which I will be on campus and teach two courses. In Fall 2024, I am at the Simons Institute for the Theory Of Computing as a Visiting Scientist.
Publications by Year|Teaching|Students|Research Statement
email: szhou@ucr.edu
Selected Publications
Preprints
Debiasing and a local analysis for population clustering using semidefinite programming. S. Zhou. 2024. This paper supersedes an earlier version dated Nov. 2023. pdf
Semidefinite programming on population clustering: a global analysis. S. Zhou. 2023. pdf
Sharper rates of convergence for the tensor graphical Lasso estimator. S. Zhou and K. Greenewald. To appear in Proceedings of 2024 IEEE International Symposium on Information Theory (ISIT 2024), Athens, Greece. pdf / link to the conference proceedings
Thresholded Lasso for high dimensional variable selection. S. Zhou. 2023. Preliminary results appeared in Advances in Neural Information Processing Systems 22, pp 2304--2312. pdf
Tensor models for linguistics pitch curve data of native speakers of Afrikaans. M. Hornstein, S. Zhou, K. Shedden. pdf
Journal Papers
Concentration of measure bounds for matrix-variate data with missing values. S. Zhou. Bernoulli 30(1), pp 198–226, 2024. pdf
-- Originally posted on arXiv 2008.03244 titled "The tensor quadratic forms" (71 Pages), August, 2020
The tensor graphical Lasso (TeraLasso). K. Greenewald, S. Zhou, A. Hero. Journal of Royal Statistical Society, Series B 81 (5), pp 901--931, 2019. pdf /Journal Link
Joint mean and covariance estimation with unreplicated matrix-variate data. M. Hornstein, R. Fan, K. Shedden, S. Zhou. Journal of the American Statistical Association, Vol. 114 (526), pp 682--696, 2019. pdf / R package
Sparse Hanson-Wright inequalities for subgaussian quadratic forms. S. Zhou. Bernoulli, Vol. 25 (3), pp 1603--1639, 2019. pdf
Errors-in-variables models with dependent measurements. M Rudelson, S. Zhou. Electronic Journal of Statistics, Vol. 11 (1), pp 1699--1797, 2017. pdf
On hierarchical routing in doubling metrics. with T.H. Chan, A. Gupta, B.M. Maggs. ACM Transactions on Algorithms (TALG) Vol. 12 (4), pp 1--22, 2016. pdf
Gemini: Graph estimation with matrix variate normal instances. S. Zhou. Annals of Statistics, Vol. 42 (2), pp 532--562, 2014. Journal Link
Reconstruction from anisotropic random measurements. M. Rudelson, S. Zhou. IEEE Transactions on Information Theory, Vol. 59. (6), pp 3434--3447, 2013. pdf / talk slides
On convergence of Kronecker graphical Lasso algorithms. T. Tsiligkaridis, A. Hero and S. Zhou. IEEE Transactions on Signal Processing ,Vol. 61 (7), pp 1743–1755, 2013. pdf
High-dimensional covariance estimation based on Gaussian graphical models. S. Zhou, P. Rutimann, M. Xu, P. Buhlmann. Journal of Machine Learning Research, Vol. 12 (91), pp 2975--3026, 2011. Journal Link / pdf
The adaptive and the thresholded Lasso for potentially misspecified models (and a lower bound for the Lasso), S. van de Geer, P. Buhlmann, S. Zhou. Electronic Journal of Statistics 2011, Vol. 5, 688-749. Journal Link
A Statistical framework for differential privacy. L. Wasserman, S. Zhou. Journal of the American Statistical Association, Vol. 105 (489), pp 375--389, 2010 (Appeared as a featured article). pdf
Edge disjoint paths in moderately connected graphs. S. Rao, S. Zhou. SIAM Journal on Computing, Vol. 39 (5), pp 1856--1887, 2010. Journal version
Time varying undirected graphs. S. Zhou, J. Lafferty, L. Wasserman. Machine Learning Journal, Vol 80, Numbers 2--3, pp 295--319, 2010 (Invited and peer reviewed). pdf
Compressed and privacy sensitive sparse regression. S. Zhou, J. Lafferty, L. Wasserman. IEEE Transactions on Information Theory, Vol. 55 (2), pp 846--866, 2009. pdf
Separating populations with wide data: a spectral analysis. with A. Blum, A. Coja-Oghlan, A. Frieze. Electronic Journal of Statistics, Vol. 3, pp 76--113, 2009. Journal Link/ pdf
Conference Papers
Differential privacy with compression. S. Zhou, K. Ligget and L. Wasserman. In Proceedings of the 2009 IEEE International Symposium on Information Theory (ISIT), 2009.
Thresholding procedures for high dimensional variable selection and statistical estimation. S. Zhou. Advances in Neural Information Processing Systems 22, pp 2304--2312, 2009. pdf Full version: arXiv 1002.1583
The Bigraphical Lasso. A. Kalaitzis, J. Lafferty, N. D. Lawrence and S. Zhou. In Proceedings of the 30th International Conference on Machine Learning (ICML), PMLR 28(3):1229–1237, 2013.
Time-dependent spatially varying graphical models, with application to brain fMRI data analysis. K. Greenewald, S. Park, S. Zhou, A. Giessing. Advances in Neural Information Processing Systems 30, 2017. pdf
Precision matrix estimation with noisy and missing data. R. Fan, B. Jang, Y. Sun, and S. Zhou. In Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 89:2810–2819, 2019.
Selected Talks
Tensor graphical models for complex and high dimensional data. Algorithmic Advances for Statistical Inference with Combinatorial Structure, Simons Institute for the Theory of Computing, Berkeley, CA, Oct. 11-15, 2021 abstract, video
Reconstruction from anisotropic random measurements. Coding, Complexity, and Sparsity Workshop (SPARC), University of Michigan, Ann Arbor, MI, August 5-7, 2013 slides
High-dimensional covariance estimation based on Gaussian graphical models. IMA workshop on High Dimensional Phenomena, University of Minnesota, Minneapolis, MN, September 2011 slides
Fall in love with the journey -- Jared L. Cohon