Spike propagation in a nanolaser-based optoelectronic neuron

Abstract

With the recent growth of Artificial Intelligence and Neural Networks systems, alternatives to the Von Neumann architecture are on development to run these algorithms efficiently in terms of speed, size and power consumption [1] – [3] . Electrical transmission of information is highly dissipative, transistors are reaching their size limit since their functionalities are affected by quantum effects and digital computing is falling short to manage the vast number of continuous variables involved in AI algorithms at a fast pace.