posted on 2015-07-22, 16:29authored byThomas Erlebach, Matthew Radoja
In a capacitated network design game, each of n players selects
a path from her source to her sink. The cost of each edge is shared
equally among the players using the edge. Every edge has a finite capacity
that limits the number of players using the edge. We study the
price of stability for such games with respect to the max-cost objective,
i.e., the maximum cost paid by any player. We show that the price of
stability is O(n) for symmetric games, and this bound is tight. Furthermore,
we show that the price of stability for asymmetric games can be
Ω(n log n), matching the previously known upper bound. We also prove
that the convergence time of best response dynamics cannot be bounded
by any function of n.
History
Citation
Proceedings of the 8th International Symposium on Algorithmic Game Theory (SAGT 2015), Lecture Notes in Computer Science 9347, pp. 57–68, 2015
Author affiliation
/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Computer Science
Source
SAGT 2015, Saarbrücken
Version
AM (Accepted Manuscript)
Published in
Proceedings of the 8th International Symposium on Algorithmic Game Theory (SAGT 2015)