Constraint-based optimization and utility elicitation using the minimax decision criterion
Artificial Intelligence
Regret-based incremental partial revelation mechanisms
AAAI'06 Proceedings of the 21st national conference on Artificial intelligence - Volume 1
Auctions with severely bounded communication
Journal of Artificial Intelligence Research
Mechanism design with partial revelation
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Automated design of multistage mechanisms
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Complexity of mechanism design
UAI'02 Proceedings of the Eighteenth conference on Uncertainty in artificial intelligence
Mechanisms for information elicitation
Artificial Intelligence
Simplicity-expressiveness tradeoffs in mechanism design
Proceedings of the 12th ACM conference on Electronic commerce
Hi-index | 0.00 |
In most mechanism design settings, optimal general-purpose mechanisms are not known. Thus the automated design of mechanisms tailored to specific instances of a decision scenario is an important problem. Existing techniques for automated mechanism design (AMD) require the revelation of full utility information from agents, which can be very difficult in practice. In this work, we study the automated design of mechanisms that only require partial revelation of utilities. Each agent's type space is partitioned into a finite set of partial types, and agents (should) report the partial type within which their full type lies. We provide a set of optimization routines that can be combined to address the trade-offs between the amount of communication, approximation of incentive properties, and objective value achieved by a mechanism. This allows for the automated design of partial revelation mechanisms with worst-case guarantees on incentive properties for any objective function (revenue, social welfare, etc.).