FPGA implementation of Hopfield neural network via simultaneous perturbation rule
M. Wakamura, Yutaka Maeda · Society of Instrument and Control Engineers of Japan · 2003
Hopfield neural network (HNN) is a typical example of recurrent neural networks. Software implementation of HNN does not obtain sufficient speed in the operation. Therefore, hardware implementation, especially, FPGA implementation of HNN is very promising. Originally, the weights of HNN are calculated by patterns to be memorized. However, we adopt a recursive learning method via the simultaneous perturbation learning rule.