Yisong Yue, PhD
Professor of Computing and Mathematical Sciences,
Director of DOLCIT
California Institute of Technology
Professor Yisong Yue is a professor of Computing and Mathematical Sciences at the California Institute of Technology (Caltech). He is the director of the DOLCIT, which is broadly centered around research pertaining statistical decision theory, statistical machine learning, and optimization. He is also on the Scientific Committee for Caltech's new Center for Autonomous Systems and Technology (CAST).
His core interest is in developing practical theory of machine learning that pushes principled algorithm design towards real-world applications. Professor Yue works closely with domain experts to understand the frontier challenges in applied machine learning, distill those challenges into mathematically precise formulations, and develop novel methods to tackle them.
Professor Yue's fundamental research is largely focused on the following areas:
Interactive Machine Learning, which pertains to settings where the system must repeatedly interact with the environment (which might be a human), and possibly learn from collecting data through interactions. Includes settings such as active learning, Bayesian optimization, bandits, reinforcement learning, and experiment design.
Structured Machine Learning, which pertains to settings where we are modeling complex phenomena whose ambient representation is exponentially or infinitely large. Structure refers to mathematical abstractions that constrain the space to be more tractable both statistically and computationally. Common research directions include integrating learning with other domains such as physics, control, and logic, as well as unifying the respective theories.
Education & Training
PhD: Cornell University (2011)
BS: University of Illinois at Urbana-Champaign (2005)