Symbolic Systems Forum - Yoav Shoham, Computer Science Department, "What We Talk About When We Talk About Multi-Agent Learning"

Yoav Shoham, Computer Science Department, "What We Talk About When We Talk About Multi-Agent Learning"

http://cs.stanford.edu/~shoham

ABSTRACT:

Learning in multi-agent systems has a several-decade history

within game theory, and a recent surge of interest within computer science

(and, in particular, AI). When one looks closely, it becomes apparent

that the very questions addressed in this area are often unclear, or at

the very least only implicit. I will identify (based on joint work with

Teg Grenager and Rob Powers) five distinct agendas one can define in the

context of multi-agent learning, and then describe (based on joint work

with Rob Powers) some recent progress in advancing one of these, the

agent-centric agenda.

BIO:

Yoav Shoham is a Professor of Computer Science. He received his PhD at

Yale University, and did post-doctoral research at the Weizmann Institute

in Rehovot, Israel. His research includes artificial intelligence work in

logic and multi-agent systems, as well as game theory and electronic

commerce.