Fast Algorithms for Game-Theoretic Centrality Measures

12/06/2015 ∙ by Piotr Lech Szczepański, et al. ∙ 0

In this dissertation, we analyze the computational properties of game-theoretic centrality measures. The key idea behind game-theoretic approach to network analysis is to treat nodes as players in a cooperative game, where the value of each coalition of nodes is determined by certain graph properties. Next, the centrality of any individual node is determined by a chosen game-theoretic solution concept (notably, the Shapley value) in the same way as the payoff of a player in a cooperative game. On one hand, the advantage of game-theoretic centrality measures is that nodes are ranked not only according to their individual roles but also according to how they contribute to the role played by all possible subsets of nodes. On the other hand, the disadvantage is that the game-theoretic solution concepts are typically computationally challenging. The main contribution of this dissertation is that we show that a wide variety of game-theoretic solution concepts on networks can be computed in polynomial time. Our focus is on centralities based on the Shapley value and its various extensions, such as the Semivalues and Coalitional Semivalues. Furthermore, we prove #P-hardness of computing the Shapley value in connectivity games and propose an algorithm to compute it. Finally, we analyse computational properties of generalized version of cooperative games in which order of player matters. We propose a new representation for such games, called generalized marginal contribution networks, that allows for polynomial computation in the size of the representation of two dedicated extensions of the Shapley value to this class of games.



There are no comments yet.


page 1

page 2

page 3

page 4

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.