Wenhao Pan

I am a third-year Statistics PhD student in the Department of Statistics at University of Washington advised by Kevin Jamieson from the Paul G. Allen School of Computer Science & Engineering. I am also a part-time Applied Scientist Intern at the SCOT - Forecasting Science Department of Amazon. Before that, I completed my BA in Computer Science and BA in Statistics at the University of California, Berkeley, graduating Summa cum laude. At Berkeley, I had the privilege of working with Anil Aswani, Cyrus Dioun, and Haichen Wang, focusing on the application of optimization and deep learning models in sociology, particle physics, and tensor completion.

My research interests lie at the intersection of machine learning and mathematical optimization, with a particular focus on decision-making, sampling, and prediction problems involving sequential data. My theoretical research, advised by Kevin Jamieson, centers on developing provably optimal estimator for the low-rank matrix trace regression under the random design. One key application is to enhnace collaborative filtering based recommender systems by designing matrix completion bandit algorithms. On the applied side, my work as an Applied Scientist Intern at Amazon focuses on improving foundation models for time series forecasting, specifically by integrating classical statistical models to boost their zero-shot prediction capabilities.

My protrait on the right was taken by my good friend Li at Lake Tapps Park in Bonney Lake, WA.