An optimally robust, fast-learning, pattern recognizer derived from a noniterative neural network learning theory

It is proved analytically that, whenever the input-output mapping of a one-layered, hard-limited perceptron satisfies a positive, linear independency (PLI) condition, the connection matrix A to meet this mapping can be obtained noniteratively in one step from an algebraic matrix equation containing an N/spl times/M input matrix U. Each column of U is a given standard pattern vector, and there are M standard patterns to be classified. It is also analytically proved that sorting out all nonsingular submatrices U/sup k/ in U can be used as an automatic feature extraction process in this noniterative-learning system. This paper reports the theory, the design, and the experiments of a superfast-learning, optimally-robust, neural network pattern recognition system derived from this novel noniterative learning theory. An unedited video movie showing the speed of learning and the robustness in recognition of this novel pattern recognition system is demonstrated. Comparison to other neural network pattern recognition and feature extraction systems are discussed.