A learning scheme for hardware implementation of feedforward neural networks

A learning scheme is proposed for hardware implementation of single-pattern and multiple-pattern learning of feedforward neural networks (FNNs). The proposed learning scheme is quite different from that for the conventional hardware implementation but quite similar to the conventional software learning approach. The proposed learning scheme is implemented by inserting switching circuits between multi-layered feedforward circuitry and learning circuitry. Learning circuitry of feedforward neural networks (FNNs) are implemented by employing MEBP (Modified Error Back-Propagation) learning rule. The proposed scheme has been implemented using conventional CMOS technology and its operation has been verified using HSPICE circuit simulator. The implemented FNNs with learning produce the output voltage, which is uniquely determined by any pair of learning input patterns. The proposed learning scheme is very suitable for the future implementation of the large-scale neural networks with learning.