|
||||
|
|
|
|
|
Associate Professor
 |
Department of Computer Science
Johns Hopkins University 3400 N. Charles Street, NEB 224 Baltimore, MD 21218-2691 U.S.A. Email: jason@cs.jhu.edu
Web: http://cs.jhu.edu/~jason
Office: Computational Science and Engineering Building 324C Phone: (410) 516-8438 (dial 516-THETA) Fax: (410) 516-6134 |
| Department of Computer Science | (my primary appointment) |
| Center for Language and Speech Processing | (my major multi-departmental center at JHU) |
| Human Language Technology Center of Excellence | (another large group I'm involved with at JHU) |
| Department of Cognitive Science | (my joint appointment) |
Computers now have access to lots of natural language data (text and speech). As a computational linguist, I help computers learn to understand human language.
I design algorithms for applying and learning
statistical models
that exploit linguistic structure
to improve performance on real data.
A central theme in my work is structured prediction -- learning to predict many interrelated variables at once, as in parsing, machine translation, morphology, phonology, and information extraction. To this end, my students and I have made numerous algorithmically novel contributions to dynamic programming, belief propagation, finite-state methods, variational inference, and semisupervised learning. We have been developing a high-level declarative programming language, Dyna, that encapsulates many practical efficiency tricks and makes it easy to experiment with new algorithms and models.
See my research summary for more information!
Undergraduates are often curious about their teachers' secret lives. In the name of encouraging curiosity-driven research, here are a few photos:
And some non-photos:
If I had a geek code, it would be GCS/O/M/MU d-(+) s:- a C++$ ULS+(++) L++ P++ E++>+++ W++ N++ o+ K++ w@ !O V- PS++ PE- Y+ PGP b++>+++ !tv G e++++ h- r+++ y+++, but I disapprove of the feeping creaturism of these things.
