Stanislav Minsker - Publications


  • The Geometric Median and Applications to Robust Mean Estimation (with N. Strawn, 2023).

  • Efficient median of means estimator. In COLT 2023, 36th Conference on Learning Theory.

  • Robust and tuning-free sparse linear regression via square-root Slope (with M. Ndaoud and L. Wang, 2022).

  • Median of means principle for Bayesian inference (with S. Yao, 2022).

  • U-statistics of growing order and sub-Gaussian mean estimators with sharp constants. To appear in Mathematical Statistics and Learning (2023).

  • Minimax Supervised Clustering in the Anisotropic Gaussian Mixture Model: A new take on Robust Interpolation (with M. Ndaoud and Y. Shen, 2021).

  • Robust estimation of covariance matrices: adversarial contamination and beyond (with L. Wang). To appear in Statistica Sinica (2023).

  • Robust and efficient mean estimation: approach based on the properties of self-normalized sums (with M. Ndaoud). Electronic Journal of Statistics, Vol. 15(2), 6036-6070. (2021).

  • Asymptotic normality of robust risk minimizers. Submitted (2020).

  • Robust modifications of U-statistics and applications to covariance estimation problems (with X. Wei). Bernoulli 26, No. 1, 694-727 (2020).

  • Excess risk bounds in robust empirical risk minimization (with T. Matthieu). Information and Inference: A Journal of the IMA, Vol. 10(4), 1423–1490 (2021).

  • Uniform bounds for robust mean estimators. Submitted (2019).

  • User-friendly covariance estimation for heavy-tailed distributions: a survey and recent results (with Yuan Ke, Zhao Ren, Qiang Sun and Wen-Xin Zhou). Statistical Science, vol. 34, No. 3, 454-471 (2019).

  • Moment inequalities for matrix-valued U-statistics of order 2 (with X. Wei). Electronic Journal of Probability 24, No. 133, 1–32. (2019).

  • Distributed statistical estimation and rates of convergence in normal approximation. Electronic Journal of Statistics, Vol. 13(2), 5213-5252. (2019).
    PDF of the most recent version; also see arXiv:1704.02658 for an older version of the paper.

  • Structured signal recovery from non-linear and heavy-tailed measurements (with L. Goldstein and X. Wei). IEEE Transactions on Information Theory,Vol. 64, Iss. 8, pages 5513 - 5530 (2018).

  • Sub-Gaussian estimators of the mean of a random matrix with heavy-tailed entries. Annals of Statistics, Vol. 46, pages 2871-2903 (2018).

  • Estimation of the covariance structure of heavy-tailed distributions (with X. Wei). Neural Information Processing Systems (NeurIPS) 2017.

  • Multiscale Dictionary Learning: Non-Asymptotic Bounds and Robustness (with M. Maggioni and N. Strawn). Journal of Machine Learning Research, 17(2):1−51 (2016).
    Abstract and PDF.
    Code on M. Maggioni’s webpage.

  • Geometric Median and Robust Estimation in Banach Spaces. Bernoulli 21, no. 4, 2308–2335 (2015).

  • On some extensions of Bernstein's inequality for self-adjoint operators. Statistics and Probability Letters, 127, p. 111-119 (2017).

  • Functional linear model with subgaussian design: L_1 - penalization approach (with V. Koltchinskii). Journal de l’Ecole Polytechnique – Mathematiques, 1, p. 269-330 (2014).

  • Robust and scalable Bayes via a median of subset posterior measures (with S. Srivastava, L. Lin and D. Dunson). Journal of Machine Learning Research, 18(124):1−40 (2017).
    M-posterior code is available here; R package can be downloaded here.

  • Active Learning for Personalized Medicine (with G. Cheng and Y. Zhao). Journal of the American Statistical Association, vol. 111, number 514, p. 875-887 (2015).

  • Scalable and Robust Bayesian Inference via the Median Posterior (with S. Srivastava, L. Lin and D. Dunson). In ICML 2014, 31st International Conference on Machine Learning.

  • Multiscale Dictionary and Manifold Learning: Non-Asymptotic Bounds for the Geometric Multi-Resolution Analysis (with M. Maggioni and N. Strawn). In iTWIST 2014, International Traveling Workshop on Interactions between Space Models and Technology.
    Conference Proceedings.

  • Learning Extreme Values and Associated Level Sets of a Regression Function via Selective Sampling. In COLT 2013, 26th Conference on Learning Theory.
    Abstract and PDF.

  • Plug-in approach to Active Learning. Journal of Machine Learning Research 13, p. 67-90 (2012).
    Abstract and PDF.
    Improved version of some bounds can be found in Chapter 2 of my Ph.D. thesis.

  • Sparse recovery in convex hulls of infinite dictionaries (with V. Koltchinskii). In COLT 2010, 23rd Conference on Learning Theory.
    Extended version is included in Chapter 3 of my Ph.D. thesis.

Ph.D. Thesis

Non-Asymptotic Bounds for Prediction Problems and Density Estimation. Georgia Institute of Technology, 2012. PDF.