Cooperative Games with Coalition Structures

Cooperative Games with Coalition Structures

1967 | R. J. Aumann, Jerusalem; J. H. Dreze, Louvain
Aumann and Dreze examine cooperative games with coalition structures, showing how various solution concepts can be defined relative to a coalition structure. They establish theorems connecting solution notions defined for a coalition structure B with those defined for each coalition in B. These solution concepts include the kernel, nucleolus, bargaining set, value, core, and von Neumann-Morgenstern solution. Interestingly, a single function, $ v_{x}^{*} $, plays a central role in five of these six solution concepts, despite their differences. This function naturally appears in all five cases, highlighting a common game-theoretic phenomenon that transcends specific solution notions. The paper defines a game as a pair (N, v), where N is the set of players and v is a function on subsets of N. A payoff vector is a function x on N, and a constrained game is a triple (N, v, X). The paper discusses how solution concepts can be defined for games with arbitrary coalition structures, and how these relate to solutions defined on individual coalitions. It also explores the core of a game with a coalition structure and shows that it equals the core of the superadditive cover of the game. Examples are provided to illustrate the results, and the paper concludes with a discussion of the rationale for studying games with coalition structures. The authors acknowledge that their analysis is not comprehensive, as some important solution concepts are not covered. They also note that the numbering system is keyed to the sections, with theorems and corollaries named accordingly. The paper is a significant contribution to game theory, demonstrating the importance of coalition structures in defining and analyzing solution concepts.Aumann and Dreze examine cooperative games with coalition structures, showing how various solution concepts can be defined relative to a coalition structure. They establish theorems connecting solution notions defined for a coalition structure B with those defined for each coalition in B. These solution concepts include the kernel, nucleolus, bargaining set, value, core, and von Neumann-Morgenstern solution. Interestingly, a single function, $ v_{x}^{*} $, plays a central role in five of these six solution concepts, despite their differences. This function naturally appears in all five cases, highlighting a common game-theoretic phenomenon that transcends specific solution notions. The paper defines a game as a pair (N, v), where N is the set of players and v is a function on subsets of N. A payoff vector is a function x on N, and a constrained game is a triple (N, v, X). The paper discusses how solution concepts can be defined for games with arbitrary coalition structures, and how these relate to solutions defined on individual coalitions. It also explores the core of a game with a coalition structure and shows that it equals the core of the superadditive cover of the game. Examples are provided to illustrate the results, and the paper concludes with a discussion of the rationale for studying games with coalition structures. The authors acknowledge that their analysis is not comprehensive, as some important solution concepts are not covered. They also note that the numbering system is keyed to the sections, with theorems and corollaries named accordingly. The paper is a significant contribution to game theory, demonstrating the importance of coalition structures in defining and analyzing solution concepts.
Reach us at info@futurestudyspace.com