Assistant Professor of Computer Science and Engineering
Probabilistic categorial grammar induction with higher-order unification
Recent work has demonstrated effective learning algorithms for a
variety of semantic parsing problems, where the goal is to
automatically recover the underlying meaning of input sentences.
Although these algorithms can work well, there are often designed
only for specific languages or meaning representations.
In this talk, I will describe new, probabilistic CCG grammar
induction algorithms that are largely language and representation
independent. The key idea is to use labeled meaning representations
to guide the induction of syntactic categories, via higher-order
This method, when combined with new probabilistic models for
categorial lexicon generalization, provides highly accurate semantic
parsing results, with significantly reduced engineering effort. I will
also briefly sketch future directions, including our efforts to use semantic
parsing for grounded language understanding and in conversational
Friday, October 14, 2011
Paccar Hall Room 291
Reception to Follow
The University of Washington is committed to providing access, equal
opportunity, and reasonable accommodations in its services, programs, activities, education, and
employment for individuals with disabilities.
To request disability accommodations, please contact the Office of the ADA
Coordinator in advance.
545-6450 (voice); 543-6452 (TDD); email@example.com (e-mail).