Unsupervised Learning Architecture Based on Spike-Timing-Dependent Plasticity Using Flash Memory Synaptic Devices

In this paper, we propose a hardware-based neural network architecture to implement pulse scheme required for unsupervised learning using flash memory synaptic devices. We perform circuit simulation to verify the systematic operation of multiple neuron system, and calculate energy consumption during synaptic weight changes. Also, the recognition accuracy is obtained in MNIST handwritten digit pattern learning by applying spike-timing-dependent plasticity rule through system level simulation.