The original formulation of SharedPlans (Grosz and Sidner, 1990a) was developed to provide a model of collaborative planning in which it was not necessary for one agent to have intentions-to toward an act of a different agent. Unlike other contemporaneous approaches (Searle, 1990), this formulation provided for two agents to coordinate their activities without introducing any notion of irreducible joint intentions. However, it only treated activities that directly decomposed into single-agent actions, did not address the need for agents to commit to their joint activity, and did not adequately deal with agents having only partial knowledge of the way in which to perform an action. This paper provides a revised and expanded version of SharedPlans that addresses these shortcomings. It also reformulates Pollack's definition of individual plans (Pollack, 1990) to handle cases in which a single agent has only partial knowledge; this reformulation meshes with the definition of Shar...

Automated intelligent agents inhabiting a shared environmentmust coordinate their activities. Cooperation { not merely coordination { may improve the performance of the individual agents or the overall behavior of the system they form. Research in Distributed Arti cial Intelligence (DAI) addresses the problem of designing automated intelligent systems which interact e ectively. DAI is not the only eld to take on the challenge of understanding cooperation and coordination. There are a variety of other multi-entity environments in which the entities coordinate their activity and cooperate. Among them are groups of people, animals, particles, and computers. We argue that in order to address the challenge of building coordinated and collaborated intelligent agents, it is bene cial to combine AI techniques with methods and techniques from a range of multi-entity elds, such as game theory, operations research, physics and philosophy. To support this claim, we describe some of our projects, where we have successfully taken an interdisciplinary approach. We demonstrate the bene ts in applying multi-entity methodologies and show the adaptations, modi cations and extensions necessary for solving the DAI problems.

Abstract Negotiations are very important in a multi-agent environment, particularly, in an environment where there are conflicts between the agents, and cooperation would be beneficial. We have developed a general structure for a Negotiating Automated Agent that consists of five modules: a Prime Minister, a Ministry of Defense, a Foreign Office, a Headquarters and Intelligence. These modules are implemented using a dynamic set of local-agents belonging to the different modules. We used this structure to develop a Diplomacy player, Diplomat. Playing Diplomacy involves a certain amount of technical skills as in other board games, but the capacity to negotiate, explain, convince, promise, keep promises or break them, is an essential ingredient in good play. Diplomat was evaluated and consistently played better than human players.

Abstract. This paper presents some of the key techniques for reaching agreements in multi-agent environments. It discusses game-theory and economics based techniques: strategic negotiation, auctions, coalition formation, market-oriented programming and contracting. It also presents logical based mechanisms for argumentations. The focus of the survey is on negotiation of self-interested agents, but several mechanisms for cooperative agents who need to resolve conflicts that arise from conflicting beliefs about different aspects of their environment are also mentioned. For space reasons, we couldn’t cover all the relevant works, and the papers that are mentioned only demonstrate the possible approaches. We present some of the properties of the approaches using our own previous work. 1

To my family... for their love and support. especially my husband Kunming, for taking care of our children such that I have time working on my research. to my children: Alex and Esther, wonderful gifts from God, who can To my parents... patiently and quietly listen to a paper instead of a story when they were still infants. who always encourage me and show me support when I was in the lurch. ii Acknowledgments First, I want to thank God for giving me chance to study here, helping me growing and giving me idea in research. I would like to thank each of the members of my dissertation committee for their time, and commitment as well. And I would like to thank Dr. Matthews for giving

This paper grows out of a conference on Methodology in International Relations convened at Rice University June 26-28, 1998. The conference was organized by the Division of Social Sciences and the James A. Baker III Institute for Public Policy at Rice, and the Program in Foreign Policy Decision Making at Texas A&M. The objective of the conference was to explore ways in which various methodological traditions address a substantive topic in international relations research. Research designs were to be developed and discussed at the Rice conference, followed by full-blown papers prepared for presentation at the February 1999 International Studies Association Meetings in Washington DC. Ultimately, an edited volume containing these papers is to be published. Five methodological traditions are represented among the papers: rational choice/game theory; dynamic modeling; quantitative approaches; quasi-experimental and simulation approaches; and qualitative/case study approaches. The unifying

Coalition formation in multi-agent systems is a process where agents form coalitions and work together to solve a joint problem via cooperating or coordinating their actions within each coalition. It is important for distributed applications ranging from electronic business to mobile and ubiquitous computing where adaptation to changing resources and environments is crucial. Coalition formation is useful as it may increase the ability of agents to accomplish tasks and achieve their goals. However, in complex real-world environments that agents operate in, the available resources are generally constrained. Agents only have incomplete even inaccurate information about the dynamically changing world. The occurrence of events may require the agents to react in a real-time manner. Agents’ actions may result in uncertain outcomes. These factors inevitably influence the formation process and formation outcome of a coalition. We employ a learning-based two-phased coalition formation approach to help agents form coalitions in complex environments. The approach consists of (1) a two-phase (planning and instantiation) coalition formation model, (2) a two-level (strategic and tactical) learning mechanism, (3) an adaptive, confidence-based negotiation strategy, and

In domains of limited resources, problem solving and execution of solutions may require the satisfaction of resource use constraints among a group of collaborating agents. One way for the agents to agree to the distribution of limited resources is through negotiation. In this paper we present how an agent that decides it must negotiate for the use of resources that it needs to reason or execute a task can use case-based reasoning (CBR) and utility to learn, select, and apply negotiation strategies. The negotiation process is situated in the current world description, self state, and also dynamically changing evaluation criteria and constraints. Consequently, the negotiation strategies that an agent uses vary greatly. To determine the negotiation strategy an agent uses CBR to compare the new situation to old cases (from its situated case base) and learns from its previous experiences how it should negotiate. This unique synergy allows us to address real-time resource allocation and efficient knowledge management: (1) the use of negotiation reduces communication traffic since knowledge updates and exchanges are performed only when necessary, and (2) the use of CBR and utility streamlines the decision process so that an agent can obtain a "good-enough, soon-enough" negotiation strategy effectively.

Abstract not found

Multi-agent artificial decision systems require a praxeology, or science of efficient action, that accommodates complex interactions between decision makers. Conventional praxeologies are built on the paradigm of rational choice, which comprises the two companion premises of totally-ordered preferences and individual rationality. Specifying the total orderings, however, is difficult for complex systems, due to informational limitations. Bounded rationality modifies the total-ordering premise to accommodate these limitations but continues to rely upon individual rationality. This paper provides a distinct alternative to the hyperrationality of conventional rational choice and the approximations of bounded rationality by waiving reliance on both of the rational choice premises and offering an approach to decision making that is based on a well-defined mathematical notion of satisficing, or being good enough, that permits the modeling of complex interrelationships between agents, including cooperation, unselfishness, and altruism. This new praxeology provides a constructive approach to decision system design by building upon local conditional interrelationships between decision makers which lead to emergent group and individual orderings.