Implementation of learning in continuous analog circuitry
Frank Stüpmann, S. Rode, N. Schmidt, G. Geske · 2003
The chip described below is a self-learning classifier. The decision-making function is a trainable integrated analog neural network structure. The circuit not only contains the reproduction path, but also the learning on-chip. The process of weight change is fully integrated. The backpropagation algorithm is implemented in an analog circuit.