Research

Research Interest

My current research interest includes:

• High dimensional statistics: estimation, model selection and inference for static or dynamic high dimensional structures, including sparse PCA/CCA, robust low rank matrix recovery, change point detection in high dimensional regression, etc.

• Causal inference: design and analysis of randomized experiments, complex causal mechanisms (such as interference, sorrogacy, delay and so on), novel causal methodology (such as data fusion)

• Learning and computing: statistical learning/machine learning, reinforcement learning, deep learning, GenAI

• AI policy/ethics: privacy and fairness in statistical related methodology and practice.

• Statistical application: marriage between stats and biomedical study, social science, psychology, econ, etc.

Publications/Manuscripts

Design-based causal inference

• Shi, L., Wang, J. and Ding, P. (2024) Forward selection and post-selection inference in factorial designs. Annals of Statistics. Accepted.

• Shi, L. and Li, X. (2024) Some theoretical foundations on the design and analysis of randomized experiments. Journal of Causal Inference. Accepted.

• Shi, L., Ding, P. (2022+) Berry–Esseen bounds for design-based causal inference with possibly diverging treatment levels and varying group sizes.

Bandit, adaptive experiment and reinforcement learning

• Shi, L., Wei, W. and Wang, J. (2024+) Using Surrogates in Covariate-adjusted Response-adaptive Randomized Experiments with Delayed Outcomes. NeurIPS 2024.

• Shi, L., Wang, J. and Wu, T. (2023) Statistical inference on multi-armed bandits with delayed feedback. ICML 2023.

Theory of permutation and permutation tests

• Shi, L. and Ding, P. (2024+) Asymptotic theory for the quadratic assignment procedure.

High dimension statistics

• Shi, L., Wang, G. and Zou, C. (2024) Low-Rank Matrix Estimation in the Presence of Change-Points. Journal of Machine Learning Research. [Distinguished Student Paper Award Winner for ENAR 2023]

• Cui, X., Shi, L., Zhong, W. and Zou, C. (2023) Robust high-dimensional low-rank matrix estimation: optimal rate and data-adaptive tuning. Journal of Machine Learning Research.

• Shi, L. and Zou, C. (2020) Noisy Low Rank Matrix Completion Under General Bases. Stat.

• Shi, L., Yang, H. and Xue, L. (2023+) Oracle Inequalities for Sparse Principal Component Analysis based on the Linear Manifold Approximation.

Collaborative research

• Barbehenn, A., Shi, L., Shao, J., Hoh, R., Hartig, H.M., Pae, V., Sarvadhavabhatla, S., Donaire, S., Sheikhzadeh, C., Milush, J., Laird, G.M., et al. (2024) Rapid Biphasic Decay of Intact and Defective HIV DNA Reservoir During Acute Treated HIV Disease. Nature Communications. Accepted.

• Shi, L., Pang, H., Chen, C., Zhu, J. (2024+) rdborrow: An R package for Causal Inference Incorporating External Controls in Randomized Trials with Longitudinal Outcomes.

Manuscripts

Working projects

Causal inference

• Nonparametric lower bounds for Cronbach's alpha in the presence of missingness

• Estimation and inference in bipartite experiments

• Theory of two-way independent permutation with application to multiple randomization designs

Adaptive experiments and reinforcement learning

• Adaptive experiments with delayed feedback

Application work

Collaborators

I would like to thank many amazing collaborators: Sulggi Lee, Guanghui Wang, Xiaolong Cui, Wei Zhong