The Market-based Methods Group, under the direction of Associate Professor Lyle Ungar, is no longer accepting students.
It was initially concerned with the application of electronic markets to distributed scheduling and optimization problems, that might include a both intelligent agents and people. The work is both theoretical, drawing on a rich body of theory in economics, artificial intelligence, game theory and decentralized optimization, and empirical, testing new results in simulation.
More recent work has dealt with a variety of issues arising in information economics, including modeling search and pricing in electronic markets and modeling the drivers of outsourcing.
Older results include: iBundle, a new allocatively-efficient iterative combinatorial auction, for problems where agents demand bundles of resources; and an "augmented tatonnement" design for problems in chemical plant coordination. Current projects include market-based methods for decentralized train scheduling, and work in the area of intermediaries, shopbots and pricebots for electronic commerce.
Computer Science Contact
(215) 898-8560 (phone)