We address optimal group manipulation in multi-dimensional, multi-facility location problems. We focus on two families of mechanisms, generalized median and quantile mechanisms, evaluating the difficulty of group manipulation of these mechanisms. We show that, in the case of single-facility problems, optimal group manipulation can be formulated as a linear or second-order cone program, under the L1and L2-norms, respectively, and hence can be solved in polynomial time. For multiple facilities, we show that optimal manipulation is NP-hard, but can be formulated as a mixed integer linear or second-order cone program, under the L1and L2-norms, respectively. Despite this hardness result, empirical evaluation shows that multi-facility manipulation can be computed in reasonable time with our formulations.
CITATION STYLE
Sui, X., & Boutilier, C. (2015). Optimal group manipulation in facility location problems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9346, pp. 505–520). Springer Verlag. https://doi.org/10.1007/978-3-319-23114-3_30
Mendeley helps you to discover research relevant for your work.