Search

A framework is proposed for analyzing various types of rational interaction. We consider a variety of restrictions on participants' moves; each leads to a different characterization of rational behavior. Under an assumption of common rationality, it is proven that participants will cooperate, rather than defect, in the Prisoner's Dilemma. We will follow the usual convention of representing a game as a payoff matrix. This is a unified framework for considering various types of interactions that occur without communication. Using assumptions about what types of moves other agents will make, a participant is able to reason about what constitutes rational behavior on its own part. Several of the characterizations of rationality have parallels in existing game theory literature, and lead to familiar results such as case analysis and iterated case analysis.

Abstract: "The DAITB (Distributed Artificial Intelligence TestBed) is a general testbed for Distributed Artificial Intelligence (DAI), supplying the user with an environment for simulating various DAI scenarios. The system allows the user to define the problem domain by specifying the various objects in the world, the agents with their attributes (such as their reasoning engines, communication protocols, rules, etc.), and the actions that can be performed. The system then simulates activity in the domain, through the interactions of the agents with each other and the artificial world."

An approach is offered for conflict resolution, and it is shown that even in a conflict situation, partial cooperative steps can be taken by interacting agents (that is, agents in fundamental conflict might still agree to cooperate up to a certain point). A Unified Negotiation Protocol is developed that can be used in all cases. It is shown that in certain borderline cooperative situations, a partial cooperative agreement (i.e., one that does not achieve all agents' goals) might be preferred by all agents, even though there exists a rational agreement that would achieve all their goals."

Abstract: "We present a general theory that captures the relationship between certain domains and negotiation mechanisms. The analysis makes it possible to categorize precisely the kinds of domains in which agents find themselves, and to use the category to choose appropriate negotiation mechanisms. The theory presented here both generalizes previous results, and allows agent designers to characterize new domains accurately. The analysis thus serves as a critical step in using the theory of negotiation in real-world applications. We show that in certain Task Oriented Domains, there exist distributed consensus mechanisms with simple and stable strategies that lead to efficient outcomes, even when agents have incomplete information about their environment. We also present additional novel results, in particular that in concave domains using all-or-nothing deals, no lying by an agent can be beneficial, and that in sub-additive domains, there often exist beneficial decoy lies that do not require full information regarding the other agent's goals."