CLASSICAL-CONDITIONING WITH PULSED INTEGRATED NEURAL NETWORKS - CIRCUITS AND SYSTEM

In this paper, we investigate on-chip learning for pulsed, integrated neural networks, We discuss the implementational problems the technolo gy imposes on learning systems, and we find that a biologically inspir ed approach using simple circuit structures is most likely to bring su ccess. We develop a suitable learning algorithm - a continuous-time ve rsion of a temporal differential Hebbian learning algorithm for pulsed neural systems with nonlinear synapses - as well as circuits for the electronic implementation. Measurements from an experimental CMOS chip are presented, Finally, we use our test chip to solve simple classica l conditioning tasks, thus verifying the design methodologies put forw ard in the paper.