Geanakoplos Zuck This research is funded under the Special Initiative on Coordination Theory and Collaboration Technology. This is one of eleven winners under that competition. How to make decisions in the face of uncertainty is a core problem in many computer science and economic modelling problems. However, there are a number of weaknesses in the traditional approaches to modelling knowledge. Among them are assumption that there are perfect reasoners and that inferences can be based on arbitrarily deep levels of knowledge. This research on coordination theory is to develop a new notion of practical knowledge upon which to base coordinated decision making. Practical knowledge will take account of several factors which limit a reasoner's deductive ability such as computational resources available to the reasoner, imperfections in the reasoner's ability to carry out correct deductions, and faultiness and incompleteness in the data available to the reasoner. Many of these factors have been studied previously in isolation. This research extends and unifies such work in order to come up with a comprehensive theory of sound reasoning in a practical multi-agent environment.
