UC Berkeley / UC Davis Joint Colloquium 2026

Berkeley-Davis Joint Colloquium. Water Tower on left, Sather Tower on right.

Event Date

Location
Mathematical Sciences Building 1147

Speaker: Ryan Tibshirani

Professor, Department of Statistics, UC Berkeley

Title: Online Conformal Prediction, Multi-Level Quantile Tracking, and Gradient Equilibrium

Abstract: This talk is about uncertainty quantification for time series prediction. The overarching goal is to provide easy-to-use algorithms with formal guarantees. The algorithms we present build upon ideas from conformal prediction and control theory, are able to prospectively model conformal scores in an online setting, and adapt to the presence of systematic errors due to seasonality, trends, and general distribution shifts. We will then discuss an extension of these ideas to the setting of probabilistic forecasting, which is essentially a generalization of the framework to handle vector-valued predictions, i.e., predictions which take the form of a set of ordered quantile forecasts at different probability levels. Finally, we will generalize this even further to discuss an abstract property in online learning called gradient equilibrium, which encapsulates these settings, and more.

Bio: Ryan Tibshirani is a Professor in the Department of Statistics at UC Berkeley. He is also a Principal Investigator in the Delphi group. From 2011-2022, he was a faculty member in Statistics and Machine Learning at Carnegie Mellon University. He did his Ph.D. in Statistics at Stanford University (2011), with Jonathan Taylor as his thesis advisor. He did his B.S. in Mathematics at Stanford (2007).
 
Faculty website (links to UC Berkeley).

 

This talk is part of the STA 290 Seminar Series.

Tags