Brief Review of Neuron Typology and Analysis of the Use of Memristor Crossbars

Neuromorphic technologies using artificial neurons and synapses can offer a more efficient solution for executing artificial intelligence algorithms than traditional computing systems. Artificial neurons using memristors have recently been developed, but they have limited biological dynamics and cannot directly interact with artificial synapses in an integrated system. This paper aims to review the complexity levels and functions of neurons and synapses and to analyze the circuit implementation of individual types of neurons and neural networks.