Graphene/MoS2−xOx/graphene photomemristor with tunable non-volatile responsivities for neuromorphic vision processing

Abstract

Conventional artificial intelligence (AI) machine vision technology, based on the von Neumann architecture, uses separate sensing, computing, and storage units to process huge amounts of vision data generated in sensory terminals. The frequent movement of redundant data between sensors, processors and memory, however, results in high-power consumption and latency. A more efficient approach is to offload some of the memory and computational tasks to sensor elements that can perceive and process the optical signal simultaneously. Here, we proposed a non-volatile photomemristor, in which the reconfigurable responsivity can be modulated by the charge and/or photon flux through it and further stored in the device. The non-volatile photomemristor has a simple two-terminal architecture, in which photoexcited carriers and oxygen-related ions are coupled, leading to a displaced and pinched hysteresis in the current-voltage characteristics. For the first time, non-volatile photomemristors implement computationally complete logic with photoresponse-stateful operations, for which the same photomemristor serves as both a logic gate and memory, using photoresponse as a physical state variable instead of light, voltage and memresistance. The polarity reversal of photomemristors shows great potential for in-memory sensing and computing with feature extraction and image recognition for neuromorphic vision.

Optogenetics-inspired manipulation of synaptic memory using all-optically controlled memristors

“…Afterward, we tested the possibility of overwriting the digit pattern "1" with the pattern of "2". This is achieved by illuminating the pixels (5,11) with red light for erasing and the pixels (1, 2, 3, 6, 7, 8, 9, 10, 13, 14, and 15) with green light for writing. The experimental results present a clear digit pattern of "2" as shown in Fig.…”

“…For instance, the conductivity of various light-sensitive material-based memristors can be regulated by light with suitable intensity and wavelength, and used for photonic neuromorphic computing. [9][10][11][12][13][14][15] However, most of the previous optoelectronic memristors work in the hybrid optic-electric mode, where an electric field is needed for device operation. 16,17 The reversible control of the device conductance (synaptic weight) solely using light is elusive, hindering the full emulation of optogenetic technologies.…”

Computing of neuromorphic materials: an emerging approach for bioengineering solutions

“…Scaling up neuromorphic hardware systems to solve bigger, more complicated issues is one of the main objectives of. [276][277][278] The number of neurons and synapses that can currently be supported by the majority of neuromorphic hardware implementations is constrained. Scalable architectures and interconnectivity strategies are being developed by researchers to support larger networks and support more powerful calculations.…”

Broadband sensory networks with locally stored responsivities for neuromorphic machine vision

“…The emerging in-sensor computing that can perform low-level and high-level image processing tasks in sensory networks has attracted numerous attentions in recent years, and various neuromorphic devices and structures including optoelectronic synapses (6)(7)(8)(9)(10)(11) and one-photosensor-one-memristor arrays (3,12) have been developed to implement in-sensor computing. In particular, human retina-inspired retinomorphic vision sensors have demonstrated their great potential in in-sensor computing because they can constitute built-in artificial neural networks (ANNs) and implement multiply-and-accumulation (MAC) operations using tunable responsivities as the weights of ANNs (1,(13)(14)(15)(16)(17)(18). However, these retinomorphic devices still suffer from the volatile nature of their responsivities, which require a sustained gate bias to maintain (1,13,14,19,20).…”