Reference: Iwasaki, Y. & Levy, A. Y. Automated Model Selection for Simulation. 1993.
Abstract: Constructing an appropriate model is crucial in reasoning successfully about the behavior of a physical situation to answer a query. In compositional modeling, a system is provided with a library of composible pieces of knowledge about the physical world called model fragments. Its task is to select appropriate model fragments to describe the situation, either for static analysis of a single state, or for the more complicated case simulation of dynamic behavior over a sequence of states. In previous work we showed how the model construction problem in general can advantageously be formulated as a problem of reasoning about {\em relevance}. This paper presents an actual algorithm, based on relevance reasoning, for selecting model fragments efficiently for the case of simulation. We show that the algorithm produces an adequate model for a given query and moreover, it is the simplest one given the constraints in the query.
Notes: Updated February 1994.
Full paper available as ps.