Events Calendar

CS Colloquium

Speaker: Prof. Vince Conitzer, Duke University

Talk Title: Computational Methods for Acting Strategically

Abstract: Abstract:

Game theory concerns settings where multiple self-interested agents (e.g., people or software agents) interact in the same environment. It attempts to describe the actions that rational strategic agents will take. Many successful real-world applications of game theory are in the context of designing a system or mechanism, for example, the design of the auctions used by major search engines to allocate advertisement slots. Game theory can be used to optimize the design, taking the strategic behavior of the agents (bidders) into account.

However, a different type of application of game theory is to design a decision support tool for one of the agents in the game. For example, we may wish to help a security force to allocate its resources strategically to defend against an attacker. Because the details of the strategic setting will vary across time and across users, computational considerations are paramount: we need algorithms that can take arbitrary games as input. Moreover, due to the ambiguities of game theory, it is not clear that we can restrict attention to a single computational problem. For example, an algorithm for computing a single Nash equilibrium may not be satisfactory if there is a better equilibrium that we might reach, or if there is concern that the other agent will not play the same equilibrium.
In this talk, I present algorithms and complexity results for a variety of computational problems in game theory, and discuss them in the context of how they can help an agent act (more) strategically.



Biography: Vincent Conitzer is an Assistant Professor of Computer Science and Economics at Duke University. He received Ph.D. (2006) and M.S. (2003) degrees in Computer Science from Carnegie Mellon University, and an A.B. (2001) degree in Applied Mathematics from Harvard University. His research focuses on computational aspects of microeconomics, in particular game theory, mechanism design, voting/social choice, and auctions. This work uses techniques from, and includes applications to, artificial intelligence and multiagent systems. Conitzer has received a CAREER award, a Sloan fellowship, the inaugural Victor Lesser dissertation award, an honorable mention for the ACM dissertation award, and several awards for papers and service at the AAAI and AAMAS conferences.



Host: Prof. Milind Tambe

Location: Hedco Neurosciences Building (HNB) - Audi

Audiences: Everyone Is Invited

Contact: Kanak Agrawal