Application of classifier systems in improving response surface based approximations for design optimization

Emergent computing paradigms, such as genetic algorithms and neural networks have found increased use in problems of engineering design. These computational tools have been shown to be applicable in providing fast function approximations, in identifying causality in numerical data, and in the solution of generically difficult design optimization problems characterized by nonconvexities in the design space and the presence of discrete and integer design variables. Another aspect of these computational paradigms that have been lumped under the broad subject category of soft computing, is the domain of artificial intelligence, knowledge-based expert systems, and machine learning. The present paper explores the use of a machine learning paradigm, the central building blocks of which are tools, such as genetic algorithms and neural networks. Such learning systems have received some attention in the field of computer science, where they have been referred to as classifier systems; the paper discusses the significance of this approach in the problem of constructing high-quality global approximations for subsequent use in design optimization.