Summary: An Approximate Truthful Mechanism for Combinatorial Auctions
with Single Parameter Agents
Aaron Archer # Christos Papadimitriou + Kunal Talwar # ’
Eva Tardos §
Abstract
Mechanism design seeks algorithms whose inputs are pro­
vided by selfish agents who would lie if advantageous. In­
centive compatible mechanisms compel the agents to tell
the truth by making it in their self­interest to do so. Often,
as in combinatorial auctions, such mechanisms involve
the solution of NP­hard problems. Unfortunately, approx­
imation algorithms typically destroy incentive compatibil­
ity. Randomized rounding is a commonly used technique
for designing approximation algorithms. We devise a ver­
sion of randomized rounding that is incentive compatible,
giving a truthful mechanism for combinatorial auctions
with single parameter agents (e.g., ``single minded bid­
ders'') that approximately maximizes the social value of
the auction. We discuss two orthogonal notions of truth­
fulness for a randomized mechanism, truthfulness with

Source: Archer, Aaron - Algorithms and Optimization Group, AT&T Labs-Research

Collections: Computer Technologies and Information Sciences