John Duchi

John Duchi's work spans statistical learning, optimization, information theory, and computation. His driving goals include: (1) Discovering statistical learning procedures that optimally trade between scarce resources--computation, communication, privacy provided to study participants--while maintaining good statistical performance. (2) Building efficient and robust large-scale optimization methods that address the spectrum of optimization, machine learning, and data analysis problems we face, allowing us to move beyond brittle and bespoke solutions. (3) Developing tools to assess and guarantee the validity of--and confidence we should have in--machine-learned systems.