Event-activated biological-inspired subwavelength (sub-λ) optical neural networks are of paramount importance for energy-efficient and high-bandwidth artificial intelligence (AI) systems. Despite the significant advances to build active optical artificial neurons using for example phase-change materials, lasers, photodetectors, and modulators, miniaturized integrated sources and detectors suited for few-photon spike-based operation and of interest for neuromorphic optical computing are still lacking. In this invited paper we outline the main challenges, opportunities, and recent results towards the development of interconnected neuromorphic nanoscale light-emitting diodes (nanoLEDs) as key-enabling artificial spiking neuron circuits in photonic neural networks. This method of spike generation in neuromorphic nanoLEDs paves the way for sub-λ incoherent neural circuits for fast and efficient asynchronous brain-inspired computation.
We study in this paper the dynamics of quantum nanoelectronic resonant tunneling diodes (RTDs) as excitable neuromorphic spike generators. We disclose the mechanisms by which the RTD creates excitable all-or-nothing spikes and we identify a regime of bursting in which the RTD emits a random number of closely packed spikes. The control of the latter is paramount for applications in event-activated neuromorphic sensing and computing. Finally, we discuss a regime of multistability in which the RTD behaves as a memory. Our results can be extended to other devices exhibiting negative differential conductance.
With the recent development of artificial intelligence and deep neural networks, alternatives to the Von Neumann architecture are in demand to run these algorithms efficiently in terms of speed, power and component size. In this theoretical study, a neuromorphic, optoelectronic nanopillar metal-cavity consisting of a resonant tunneling diode (RTD) and a nanolaser diode (LD) is demonstrated as an excitable pulse generator. With the proper configuration, the RTD behaves as an excitable system while the LD translates its electronic output into optical pulses, which can be interpreted as bits of information. The optical pulses are characterized in terms of their width, amplitude, response delay, distortion and jitter times. Finally, two RTD-LD units are integrated via a photodetector and their feasibility to generate and propagate optical pulses is demonstrated. Given its low energy consumption per pulse and high spiking rate, this device has potential applications as building blocks in neuromorphic processors and spiking neural networks.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.