I am a second-year PhD student in Statistics and Machine Learning at Cornell University, advised by Martin Wells and Yuchen Wu.
I am interested in statistical machine learning and establishing rigorous foundations for statistical and machine learning methods, while also developing new algorithms guided by theoretical insights. I am fortunate to have been supported by a Cornell University Fellowship.
Before Cornell, I was an undergraduate at UC Berkeley studying Computer Science, where I was advised by Iain Carmichael and Ryan Tibshirani.
If you're an undergrad or Master's student at Cornell and are interested in collaborating, please reach out!
Theory and methods in high-dimensional statistics, robust and stochastic optimization, reinforcement learning, language and diffusion models, sampling, and differential privacy.
Under mild smoothness, we show that controlling per-step cross-entropy (conditional KL) also controls guidance-vector error: \( \mathrm{KL} \le \varepsilon^{2} \) implies guidance MSE \( \widetilde{O}(d\varepsilon) \). This gives a sampling-error bound with a reverse log-Sobolev flavor.
ArXiv
A sharp finite-time theory showing when and why momentum fails: we decompose tracking error into initialization + noise + drift, proving momentum suppresses noise but amplifies drift with minimax dynamic-regret lower bounds.
ArXiv
A family of distributionally robust policy optimization algorithms for LLM alignment that reduce to simple relative-reward regressions, achieving minimax-optimal rates and significantly tighter constants than prior DRO-DPO methods.
ArXiv
Privatizing LSVI-UCB++ with Bernstein bonus achieves state-of-the-art regret in linear MDPs under joint differential privacy with minimal utility loss.
ArXiv
A video with Sithija Manage breaking down how I approached graduate school applications and wrote my Statement of Purpose. We discuss what to include, how to describe research experiences with technical detail, and how to tailor your SOP for committee-driven vs PI-driven admissions.
An interview about my journey through graduate school applications. We discuss my undergraduate research at Berkeley, Stanford, and USC, the admissions process, writing statements of purpose, building mentorships, and evaluating program fit.
A podcast episode on the Liftoff Podcast with Aman Manazir. I share how I got interested in competitive math, balanced internships and research at Berkeley, Stanford, and USC, and the decision between quant research and a PhD at Cornell.