Approximation in mathematical domains
Scott William Bennett · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign) · 1987
Explanation-based learning is accomplished through the generalization of an explanation produced by analysis of a single example.A theory of the domain is utilized in generating the exp anation.However, problems arise when the domain theory is intractable.Simplifications must be made in order to make the problem tractable.Well-founded simplifications based on real world nowledge are termed approximations.This paper discusses how approximation can be used to deal with the intractable domain problem in mathematical domains.The approximation method strongly supports the use of a mix of quantitative and qualitative reasoning over either a purely quantitative or qualitative approach.The approximation technique is demonstrated on one of the examples which has been implemented in the chemistry domain.