Speed up the learning process of feedforward neural networks

Among the feedforward network learning rules, the backpropagation learning rule is one that has been successfully applied to a variety of problems. However, the process of backpropagation learning is somewhat a “black box”, it suffers from several limitations. We propose a scheme which uses the Hamming coding, the partitioned network, and the logic design theory to help the feedforward network to learn so as to speed up the learning process. Our experimental results reveal that the feedforward neural networks do not need to learn blindly, in fact, they can be taught to learn. The advantages of the proposed scheme are: the network size becomes smaller, the learning rate is higher, and the learning speed is faster as well