Exciton–polariton condensates for photonic spiking neurons

Spiking neural networks are a class of artificial neural networks maintaining a strict analogy to brain-like processing. I'll show a new hardware approach in which semiconductor microcavities in strong light-matter coupling regime can operate as optical spiking neurons. We demonstrated the intrinsic property of exciton-polaritons to resemble the Leaky Integrate-and-Fire spiking mechanism. Polaritons when pumped with a pulsed laser exhibit leaky-integration due to relaxation of the excitonic reservoir, threshold-and-fire mechanism due to transition to polariton condensate, and resetting due to stimulated emission of photons. Our approach provides means for energy-efficient ultrafast processing of spike-like laser pulses at the level below 1 pJ/spike.