GROWING METHODS FOR CONSTRUCTING RECURSIVE DETERMINISTIC PERCEPTRON NEURAL NETWORKS AND KNOWLEDGE EXTRACTION

The Recursive Deterministic Perceptron (RDP) feedforward multilayer ne ural network is a generalization of the single layer perceptron topolo gy (SLPT). This new model is capable of solving any two-class classifi cation problem, as opposed to the single layer perceptron which can on ly solve classification problems dealing with linearly separable (LS) sets (two subsets X and Y of R-d are said to be linearly separable if there exists a hyperplane such that the elements of X and Y lie on the two opposite sides of R-d delimited by this hyperplane). For all clas sification problems, the construction of an RDP is done automatically and thus, the convergence to a solution is always guaranteed. We propo se three growing methods for constructing an RDP neural network. These methods perform, respectively, batch, incremental, and modular learni ng. We also show how the knowledge embedded in an RDP neural network m odel can always be expressed, transparently as a finite union of open polytopes. The combination of the decision region of RDP models, by us ing boolean operations, is also discussed. (C) 1998 Elsevier Science B .V. All rights reserved.