Abstract

This report is a summary of recent work on time-critical planning and scheduling at Brown University. Much of our research over the last six years has been concerned with systems that are capable of managing computational resources for complex problemsolving tasks, in which the time spent in decision making affects the quality of the responses generated by a system. We developed an approach to designing algorithms and allocating computational resources to algorithms that is widely cited in the areas of real-time problem solving [8, 5]. We have provided a number of basic results regarding the design of systems that manage their computational resources by using expectations about the performance of decision-making procedures and preferences over the outcomes resulting from applying those procedures [4, 6]. We have also provided a basic framework for planning systems that employ decision theoretic techniques to evaluate plans [15, 16] and to perform probabilistic prediction [12, 14, 13]. ...

Citations