BIOGRAPHY
I study how humans learn and use language from a computational perspective. The problems I study are especially interesting because the relationship between form (i.e., speech or text) and meaning is complex. Even simple sentences possess an intricate compositional structure that must be identified in order to be understood. Unfortunately, the clues to this structure are typically indirect and only partially reliable. Together with students, I am exploring computational models that exploit statistical regularities in syntax (the rule system governing sentence structure) and semantics (the system which relates to sentence meaning) in order to recover the structure and meaning of a sentence. These "structured statistical models" combine the advantages of rich compositional structures with robust processing and statistical learning. While my research is scientifically motivated, it has applications in areas such as speech recognition, information retrieval, and machine translation.
I became interested in this field because language is both easily observed and reflects the intricate structure of the human mind directly.
|
