Centre Universitaire d'Informatique

Date: Friday 26th of February 2010, 10:30 am

Location: Batelle building A, room 432-3 (3rd floor)

Tree Adjoining Grammars-based Structured Prediction Models for Parsing and Machine Translation

Speaker: Prof. Michael Collins, MIT Computer Science and Artificial Intelligence Laboratory Abstract TBA

Prof. Michael Collins research interests are in natural language processing, and machine learning. He completed a PhD in computer science from the University of Pennsylvania in 1998. After a few years at AT&T Labs-Research, he joined MIT in 2003. He is X-Consortium Associate Professor of Computer Science and a Sloan Research Fellow.

Date: Friday 26th of February 2010, 2:00 pm

Location: Battelle building A, room 432-3 (3rd floor)

Synchronous Rewriting for Natural Language Processing

Speaker: Prof. Giorgio Satta, Department of Information Engineering University of Padua

Prof. Giorgio Satta received a Ph.D. in Computer Science in 1990 from University of Padua, Italy. He was postdoctoral fellow in the Computer Science Department at the University of Pennsylvania, and is currently full professor at the University of Padua. His main research interests are in computational linguistics, mathematics of language and formal language theory. He is the current chair of the European Chapter of the Association for Computational Linguistics (EACL).

Abstract:

In synchronous rewriting two or more rewriting processes, typically context-free, can be carried out in a synchronous way. Synchronous rewriting systems are exploited in machine translation and syntax/ semantics interface, as well as in parsing applications where one needs to model syntactic structures based on discontinuous phrases or on non-projective dependency trees. In this presentation we introduce several computational problems that arise when using synchronous rewriting systems in the above mentioned applications, and present algorithms for their solutions. More specifically, we discuss an abstract framework under which parsing and translation algorithms can be unified. We also discuss the currently debated problem of the factorization of synchronous grammars.

À la Une