Machine Translation

Daniel Gildea
Department of Computer Science
University of Rochester


I will discuss the state of the art in automatic translation between human languages, showing how models of translation can be learned usding statistics over large amounts of parallel, bilingual text. The general formalism of synchronous context-free grammars, which generate stings in two languages simultaneously, can be used to describe a number of recent systems. I will present recent work on the theoretical complexity of translation with synchronous context-free grammars, as well as practical algorithms for efficient MT in this framework.


Dan Gildea is an Assistant Professor of Computer Science at the University of Rochester. His research focuses on statistical approaches to natural language processing, in particular for the tasks of machine translation and language understanding. He received his Ph.D. in computer science from UC Berkeley in 2001 and was postdoctoral scholar at U Penn for two years before coming to Rochester.

Colloquia Series page.