Research

Research Projects

Distributional Balancing Weights for Conformal Calibration under Covariate Shift

• Advisor: Prof. Guanhua Chen, University of Wisconsin–Madison
• Period: Mar. 2026 – Sep. 2026
• Keywords: Uncertainty Quantification · Distribution Shift · Conformal Prediction · Distributional Balancing

This project studies reliable uncertainty quantification under covariate shift through distributional balancing. The work develops reweighting methods for conformal calibration and investigates their connections to RKHS- and characteristic-function-based discrepancies.

Heavy-Tailed Robustness of Multiple-Knockoff Aggregation Methods

• Advisor: Prof. Lijun Wang, School of Mathematical Sciences, Zhejiang University
• Period: Sep. 2025 – Present
• Keywords: High-Dimensional Inference · Multiple Testing · Model-X Knockoffs · Heavy-Tailed Distributions

This project studies robust high-dimensional variable selection using multiple knockoffs, with a focus on false discovery control under heavy-tailed distributions and model misspecification.