Research

My research interests span a wide range of areas in Statistics and Machine Learning, including high dimensional statistical inference, dimension reduction, and causal inference. Specifically, I develop tools for uncertainty quantification of iterative algorithms (e.g. estimating prediction risks and constructing confidence intervals for parameters of interest), to improve the reliability of statistical machine learning methods.

Preprints

  1. Precise Asymptotics of Bagging Regularized M-estimators
    Takuya Koriyama, Pratik Patil, Jin-Hong Du, Kai Tan, Pierre C. Bellec (2024)

  2. Uncertainty quantification for iterative algorithms in linear models with application to early stopping
    Pierre C. Bellec, Kai Tan (2024) (alphabetical order)

  3. Fréchet sufficient variable selection with graphical structure among predictors.
    Jiaying Weng \(^*\) , Kai Tan \(^*\) (\(^*\)=equal contribution), Cheng Wang and Zhou Yu (2023)

Publications

  1. Corrected generalized cross-validation for finite ensembles of penalized estimators
    Pierre C. Bellec, Jin-Hong Du, Takuya Koriyama, Pratik Patil, Kai Tan (2023) (alphabetical order).
    Accepted, Journal of the Royal Statistical Society Series B. [Code]

  2. Multinomial Logistic Regression: Asymptotic Normality on Null Covariates in High-Dimensions
    Kai Tan, and Pierre C. Bellec (2023).
    Advances in Neural Information Processing Systems 36 (2024). [Code]

  3. Noise covariance estimation in multi-task high-dimensional linear models
    Kai Tan, Gabriel Romon, and Pierre C. Bellec (2023).
    Bernoulli. 30.3 (2024): 1695-1722. [arXiv] [slides] [Code]

  4. Variable dependent partial dimension reduction
    Lu Li, Kai Tan, Xuerong Meggie Wen, and Zhou Yu (2023) (alphabetical order).
    TEST. 32, 521–541 (2023).

  5. Sparse SIR: optimal rates and adaptive estimation
    Kai Tan, Lei Shi, and Zhou Yu (2020).
    The Annals of Statistics 48.1 (2020): 64-85.

  6. COM-negative binomial distribution: modeling overdispersion and ultrahigh zero-inflated count data
    Huiming Zhang, Kai Tan and Bo Li (2018).
    Frontiers of Mathematics in China, 13(4), 967-998.

  7. Summaries of three keynote lectures at the SAE-2018
    Kai Tan, and Lyu Ni (2018).
    Statistical Theory and Related Fields 2.2 (2018): 215-218.

  8. Interview with Professor Danny Pfeffermann
    Lyu Ni, and Kai Tan (2018).
    Statistical Theory and Related Fields 2.2 (2018): 219-221.