This paper analyzes coalitions among self-interested agents that need to solve combinatorial optimization problems to operate e ciently in the world. By colluding (coordinating their actions by solving a joint optimization prob-lem) the agents can sometimes save costs compared to operating individually. A model of bounded rationality is adopted where computation resources are costly. It is not worthwhile solving the problems optimally: solution quality is decision-theoretically traded o against computation cost. A normative, application- and protocol-independent theory of coalitions among bounded-rational agents is devised. The optimal coalition structure and its stability are signi cantly a ected by the agents ' algorithms ' performance pro les and the cost of computation. This relationship is rst analyzed theoretically. Then a domain classi cation including rational and bounded-rational agents is in-troduced. Experimental results are presented in vehicle routing with real data from ve dispatch centers. This problem is NP-complete and the instances are so large that|with current technology|any agent's rationality is bounded by computational complexity. 1

A Dissertation Presented by TUOMAS W. SANDHOLM

We investigate how delayed communication affects the dynamic formation of groups in distributed systems, where all decision-making agents join the same group because each expects to improve its own performance. For example, distributed job schedulers may form a group to utilize the idle resources of other members within the group. Forming a group is a search problem and we examine agents which use the feedback mechanism of stochastic learning automata to carry out this search. Although a group formation may have the potential for synergy, the agents must successfully coordinate their actions within the group relevant to the application. For example, job schedulers who form a group must still balance the load among the shared resources; that is, the collective actions of the schedulers need to be coordinated and greedy schedulers who all pick the same processor may not be successful. Agents may find that working alone is more desirable since their actions need not be coordinated and the r...

Real world players often increase their payoffs by voluntarily committing to play a fixed strategy, prior to the start of a strategic game. In fact, the players may further benefit from commitments that are conditional on the commitments of others. This paper proposes a model of conditional commitments that unifies earlier models while avoiding circularities that often arise in such models. A commitment folk theorem shows that the potential of voluntary conditional commitments is essentially unlimited. All feasible and individually rational payoffs of a two-person strategic game can be attained at the equilibria of one (universal) commitment game that uses simple commitment devices. The commitments are voluntary in the sense that each player maintains the option of playing the game without commitment, as originally defined.

The occurrence of cooperation poses a problem for the biological and social sciences. However, many aspects of the biological and social science literatures on this subject have developed relatively independently, with a lack of interaction. This has led to a number of misunderstandings with regard to how natural selection operates and the conditions under which cooperation can be favoured. Our aim here is to provide an accessible overview of social evolution theory and the evolutionary work on cooperation, emphasising common misconceptions.

We analyse two basic approaches of extending classical logics with

The idea that reciprocal cooperation can be obtained in a one-shot game if the players observe each other’s strategies before taking action has attracted many authors. From a strictly logical perspective, however, there cannot be such games. Nevertheless, there are games in which each player can observe which class, out of a collection of classes smaller than the number of strategies, the opponent’s strategy belongs to. For any underlying 2-player, finite, normal-form game there is a game extended with such coarsely observable strategies that has equilibria with payoffs arbitrarily close to any feasible, individually rational payoff profile. 1

The story of systems theory and cybernetics is a story of several research traditions all of which originated in the mid-twentieth century. Systems ideas emerged in a variety of locations and for different reasons. As a result the ideas were developed in relative isolation and emerged with different emphases. This paper discusses the books and people, conferences and institutes, and politics and technology that have influenced the systems movement. The schools of thought presented are general systems theory, the systems approach, operations research, system dynamics, learning organizations, total quality management, and cybernetics. Three points of view within cybernetics are discussed. Total quality management is a new addition to the list, but we feel it is appropriate because of its extensive use of systems ideas. This paper does not address artificial intelligence, complexity theory, family therapy, or other traditions which might have been included.

Our lives are built around compromise. Throughout each day we “negotiate ” with others in an attempt to obtain mutually acceptable outcomes. Drama Theory addresses problems involving multiple actors with conflicting objectives. These problems can be a regular feature of everyday life. They routinely occur at all levels of society – from personal relationships to international politics. This tutorial describes the Drama Theory paradigm and two of the techniques used to apply the theory – Confrontation Analysis and Immersive Briefings. The historical development of Drama Theory is discussed, followed by a detailed description of the ways in which an analyst can exploit Drama Theory to study various forms of conflict. Examples of the theory in use are scattered throughout the text, illustrating the application of the method within numerous and diverse domains. The goal of the tutorial is to provide the reader with sufficient knowledge of Drama Theory to recognise problems that may benefit from its application and to apply the basic techniques to their own problems.