JASA
I am a postdoc at the Wharton Department of Statistics and Data Science, advised by Yuejie Chi and Yuxin Chen. Before joining Wharton, I was hosted at Yale University for a year. Previously, I completed my PhD in Statistics at the University of Chicago with Cong Ma.
My research studies the statistical foundations of learning from heterogeneously structured data. Recent projects include low-rank matrix estimation with multi-view integration, ranking from pairwise comparisons beyond uniform sampling, and robust reinforcement learning.
News
- Sep 2026 A new preprint! It introduces bias-corrected subspace intersection (BCSI), a method for estimating subspace structure shared between two views. It improves over existing methods like AJIVE by correcting a second-order bias.
- Aug 2026 New paper on robust MDPs. We use plug-in reductions, with both span-informed and span-agnostic versions, to solve distributionally robust average-reward MDPs. We discuss their behavior compared with standard AMDPs over high- and low-tolerance regimes, and establish a matching pair of sample complexity upper and lower bounds.
Selected Papers
All papersCOLT
Preprint
Estimating shared subspace with AJIVE: the power and limitation of multiple data matrices.
Recent Papers
All papersPreprint
Bias-corrected subspace intersection: minimax-optimal shared subspace estimation in multi-view data.
Preprint
Robust average-reward Markov decision processes: minimax-optimal learning via plug-in reductions.
Preprint