We adopt the decision-theoretic principle of expected utility maximization as a paradigm for designing autonomous rational agents, and present a framework that uses this paradigm to determine the choice of coordinated action. We endow an agent with a specialized representation that captures the agent's knowledge about the environment and about the other agents, including its knowledge about their states of knowledge, which can include what they know about the other agents, and so on. This reciprocity leads to a recursive nesting of models. Our framework puts forth a representation for the recursive models and, under the assumption that the nesting of models is finite, uses dynamic programming to solve this representation for the agent's rational choice of action. Using a decision-theoretic approach, our work addresses concerns of agent decision-making about coordinated action in unpredictable situations, without imposing upon agents pre-designed prescriptions, or protocols, ...

We use the formalism of decision theory to develop principled definitions of emotional states of a rational agent. We postulate that these notions are useful for rational agent design. First, they can serve as internal states controlling the allocation of computations and time devoted to cognitive tasks under external pressures. Second, they provide a well defined implementation-independent vocabulary the agents can use to communicate their internal states to each other. Finally, they are essential during interactions with human agents in open multi-agent environments. Using decision theory to formalize the notions of emotions provides a formal bridge between the rich bodies of work in cognitive science, on the one hand, and the high-end AI architectures for designing rational artificial agents, on the other hand. 1 Introduction Our research is predicated on the thesis that concept of emotions and feelings can be formalized and be made useful in designing artificial agents t...

We use the formalism of decision theory to develop principled definitions of emotional states of a user, and we postulate that these notions should be useful in designing interactive entertainment systems.

In this paper, the potential synergy between instancebased pattern recognition and means-end (possible world) reasoning is explored, for supporting plan recognition in multi-aeroplane air-mission simulations. A combination of graph matching, induction, probabilistic principles and dynamic programming are applied to traces of aeroplane behaviour during flight manoeuvres. These satisfy the real-time constraints of the simulation. This enables the agents to recognise what other agents are doing and to abstract about their activity, at the instrumentation level. A means-end-reasoning model is then used to deliberate about and invoke standard operating procedures, based on recognised activity. The reasoning model constrains the recognition process by framing queries according to what a pilot would expect during the execution of the current plan(s). Results from experiments involving the dMARS procedural reasoning system and the CLARET pattern matching and induction system are described for ...

heir strengths and limitations. The discussion will be illustrated throughout with reference to concrete system examples. 2 Detailed Outline 2.0 Introduction and Motivation Goals Define terms and clarify the scope of the tutorial Stimulate the interest of the participants This description was written in October 2000 for submission to the IJCAI 01Tutorial Chair. Because of rapid developments in the field, the final structure of the tutorial may differ in some details from the structure described here. Questions Addressed What are "user-adaptive systems"? Why are such systems both theoretically and practically important? What can be expected in the rest of this tutorial? 2.1 Functions of User-Adaptive Systems Goals Make participants aware of all important types of useradaptive systems Introduce example systems that can be referred back to later in the tutorial Functions Discussed