Speaker: Aravind Joshi
University of Pennsylvania

Thursday, June 12, 2003
2:20pm - 3:20pm
Eucalyptus Point (Room B)

ABSTRACT

In setting up a formal system to specify a grammar formalism, the conventional (mathematical) wisdom is to start with primitives (basic primitive structures) as simple as possible and then introduce various operations for constructing more complex structures. An alternate approach is to start with complex (more complicated) primitives, which directly capture some crucial linguistic properties and then introduce some general operations for composing these complex structures. These two approaches provide different domains of locality, i.e., domain over which various types of linguistic dependencies can be specified. This latter approach, characterized as COMPLICATE LOCALLY,

SIMPLIFY GLOBALLY (CLSG), pushes non-local dependencies to become local, i.e., they arise in the basic primitive structures to start with.

The CLSG approach has led to some new insights into syntactic description, semantic composition, language generation, statistical processing, and psycholinguistic properties, all these with possible relevance to the cognitive architecture for language. I will describe these results in an introductory manner. If time permits I will briefly discuss the implications of this approach for other domains.

Bio:

Aravind K. Joshi is the Henry Salvatori Professor of Computer and Cognitive Science in the Department of Computer and Information Science and a former co-director of the Institute for Research in Cognitive Science at the University of Pennsylvania. He has worked on various areas of computational linguistics such as computational models of syntax, semantics, and discourse, mathematical linguistics, parsing, and psycholinguistic implications of processing models. He is a founding fellow of the American Association for Artificial Intelligence Science, a Fellow of the Association for Computing Machinery, and a Fellow of the Institute of Electrical and Electronic Engineers. He is a member of the National Academy of Engineering. Recently he won the David Rumelhart Prize of the Cognitive Science Society, which will be awarded in August 2003.