Recent progress in optoelectronic memristors for neuromorphic and in-memory computation

Abstract Neuromorphic computing has been gaining momentum for the past decades and appointed as replacer of outworn technology in conventional systems. Artificial neural networks (ANNs) can be composed by memristor crossbars hardware perform in-memory storage, a power, cost area efficient way. In optoelectronic memristors (OEMs), resistive switching (RS) controlled both optical electronic signals. Using light synaptic weigh modulator provides high-speed non-destructive method, not dependent on electrical wires, that solves crosstalk issues. particular, artificial visual systems, OEMs act retina combine sensing high-level image processing. Therefore, several efforts have made scientific community into developing meet demands each specific application. this review, recent advances inorganic are summarized discussed. The engineering device structure means to manipulate RS performance and, thus, comprehensive analysis is performed regarding already proposed materials their characteristics. Moreover, potential applications logic gates, ANNs more detail, systems also assessed, taking account figures merit described so far.