Research Statement of Yisong Yue

My core research interests lie in statistical machine learning, with a primary application focus in the field of information retrieval and access. In particular, I am interested in developing principled learning methods with theoretical foundations that will not only lead to practical systems of immediate benefit, but also push our ability to reason about the increasingly sophisticated information systems of the future. More broadly, I am interested in developing general learning approaches that can be applied to automate prediction tasks in a wide range of application domains. Managing digital information is a growing problem in every application domain, ranging from integrating biological data, browsing digital libraries, organizing personal content, searching in specialty domains, or filtering news feeds or Twitter updates. Current design methodologies are labor intensive and require extensive hands-on expertise. This inherently limits the scope and reasoning power of the systems that we can efficiently deploy today. Moving forward, I am particularly interested in developing and applying new machine learning approaches. Existing learning approaches have proven to be invaluable with their ability to combine coarse human feedback (e.g., “this document is relevant”) with statistical regularities of the prediction domain in order to derive effective models. This is evidenced by their widespread commercial adoption, and I am convinced that progress in machine learning will be crucial in understanding and overcoming the frontier challenges involved in developing the next generation of information access systems. In future work, I intend to build upon and extend my current line of research in exploring fundamental issues in machine learning. Two fundamental and inter-related themes in machine learning research are designing efficient methods for complex prediction domains, and leveraging new forms of training data or feedback. Broadly speaking, complex prediction refers to any task beyond simple binary classification or regression (e.g., predicting a ranking over a list of documents, a parse tree for a sentence, or an alignment between a pair of proteins). Efficient approaches typically leverage the inherent structure embedded within these prediction domains. With regards to leveraging new feedback, I am interested in developing systems that can efficiently adapt to new domains by learning through human interaction. This is an attractive approach since observed user behavior (e.g., clicks on search results, movement patterns tracked by cell phones, or behavioral patterns observed in “smart” homes) is both cheap to collect and naturally representative of the target user population (e.g., web users, individuals, or family-sized groups). My dissertation demonstrates new progress in both of these areas, and I am confident that this line of research has not only high intellectual merit, but also great potential for applications in information access and beyond.