Modified Perceptron learning rule and application abilities for Cellular Neural Networks
Keywords:
cellular neural networks, Perceptron learning rule, recurrent neural networks, Trial-Error-Correct approach
Abstract
This paper modifies the Perceptron learning rule in order to apply to all recurrent neural networks in general and cellular neural networks in particular since the original Perceptron learning rule was only used for feedforward neural networks. The idea is as follows, we link input, feedback output, and the bias of the cellular neural network to become new general input. The next step of the process can be implemented as the original Perceptron learning rule. However, cellular neural networks characterized by some features and several parameters in the learning rule that modifies the Perceptron can be added. Some examples are proposed in the paper to visualize the idea.