Computational Complexity in Mechanism Design
| Computer Science/Discrete Mathematics Seminar II | |
| Topic: | Computational Complexity in Mechanism Design |
| Speaker: | Jing Chen |
| Affiliation: | Massachusetts Institute of Technology; Member, School of Mathematics |
| Date: | Tuesday, November 27 |
| Time/Room: | 10:30am - 12:30pm/S-101 |
Some important mechanisms considered in game theory require solving optimization problems that are computationally hard. Solving these problems approximately may not help, as it may change the players rational behavior in the original mechanisms, leading to undesirable outcomes. This is particularly the case in combinatorial auctions.
Ill present special cases of combinatorial auctions where approximation algorithms do lead to mechanisms that are both computationally efficient and economically well behaved. If time allows, Ill also present conditions under which such mechanisms do not exist.