Research Interests

I am broadly interested in the problem of human and machine intelligence. To make progress, I examine this with the point of view provided by the twin windows of learning and language. In many ways, learning is the centerpiece of intelligence. This is what presumably distinguishes "intelligent" from "pre-programmed" behavior. Language is partly interesting because it is almost certainly learnt. Together they provide a testbed of challenging problems that shed insight on the nature of intelligence and if successful would lead to significant applications of the future. My specific research program therefore focuses on learning and language --- how each works in the human and how they can be replicated in a machine. (Note: I use language in the broad sense to include speech, language, and hearing that form the complete human oral communication system).

In my personal view, exciting research leading to insight and consequent innovation will arise out of the synergistic interaction of three points of view --- theory -- to provide computational and mathematical characterizations; science -- to provide an understanding of how intelligent systems work in the human; and engineering -- to translate the insights into useful real-world systems. At another level, given the unique problem of human and machine intelligence, progress will come from the union of computer science, cognitive science, and neuroscience. Consequently, I strive to keep abreast of developments in all three and strike a balance between them.

Last update: 7/05

Back to top