A MODEL OF NEURONS WITH UNIDIRECTIONAL LINEAR-RESPONSE

A model for a large network with an unidirectional linear response (UL R) is proposed in this letter. This deterministic system has powerful computing properties in very close correspondence with earlier stochas tic model based on McCulloch-Pitts neurons and graded neuron model bas ed on sigmoid input-output relation. The exclusive OR problems and oth er digital computation properties of the earlier models also are prese nt in the ULR model. Furthermore, many analog and continuous signal pr ocessing can also be performed using the simple ULR neural network. Se veral examples of the ULR neural networks for analog and continuous si gnal processing are presented and show extremely promising results in terms of performance, density and potential for analog and continuous signal processing. An algorithm for the ULR neural network is also dev eloped and used to train the ULR network for many digital and analog a s well as continuous problems successfully.