I am Fangyuan Lin and I’m a PhD student at Columbia University, Department of Statistics. I’m broadly interested in probability theory, information theory, and statistics.
I have experienced firsthand how transformative even a small act of kindness and guidance can be. The generosity of my friends and professors empowered me to grow—so I’m devoted to paying that support forward.
Education
Columbia University (Fall 2025 – Spring 2030 (Expected))
PhD in Statistics
University of California, Berkeley (2020 – 2024)
B.A. in Mathematics, Highest Honors
B.A. in Computer Science
News
- October 2026: Not All Mislabels Are Equal: Efficient Task-Aware Auditing with Confidence Sequences, with Noah Krever, Jakub Cerný, and Lily Xu. In submission; preliminary version in the NeurIPS 2026 Workshop “E-Values: From Statistics to ML”. Preprint.
- July 2026: Mitigating the Winner’s Curse While Controlling Multiplicity: e-Process Methods for Anytime-Valid Inference in Dose-Ranging Trials, with Victor K. de la Pena, Demissie Alemayehu, Victor H. de la Pena. Presented at SAVI Conference 2026. Preprint.
- June 2026: Our paper Bounded Difference Concentration for Infinitely Exchangeable Sequences with Applications to AI Benchmark Uncertainty is on arXiv.
- October 2025: Our paper An Analysis of Silk Density in Spider Webs was published in Royal Society Open Science.
- September 2025: Started my Ph.D. journey at Columbia Statistics and passed the Ph.D. Qualification Exams upon arrival.