Binary Electronic Synapses for Integrating Digital and Neuromorphic Computation in a Single Physical Platform

Abstract

As a promising advanced computation technology, the integration of digital computation with neuromorphic computation into a single physical platform holds the advantage of a precise, deterministic, fast data process as well as the advantage of a flexible, paralleled, fault-tolerant data process. Even though two-terminal memristive devices have been respectively proved as leading electronic elements for digital computation and neuromorphic computation, it is difficult to steadily maintain both sudden-state-change and gradual-state-change in a single device due to the entirely different operating mechanisms. In this work, we developed a digital–analog compatible memristive device, namely, binary electronic synapse, through realizing controllable cation drift in a memristive layer. The devices feature nonvolatile binary memory as well as artificial neuromorphic plasticity with high operation endurance. With strong nonlinearity in switching dynamics, binary switching, neuromorphic plasticity, two-dimension information store, and trainable memory can be implemented by a single device.

Self-Powered Memory Systems

“…The ability of logical judgment in the human brain is closely related to information memory, and its memory density and functional diversity are billions of times that of modern computers. The neuromorphic chip is designed to process the acquired information in a completely different way from traditional computing hardware, and by imitating the information processing structure of the human brain, it greatly improves the computer’s analytical capabilities and logical judgment capabilities. − In recent years, the miniaturization of computers has significantly improved the operating performance of traditional computing, but the continuous movement of data between the internal memory and the information processing unit consumes a lot of energy and generates excessive heat. This bottleneck limits the further improvement of the computer.…”

MXene‐ZnO Memristor for Multimodal In‐Sensor Computing

“…[1][2][3][4][5][6] In the conventional design, the analogue sensory data are initially transformed into digital data with analogue-to-digital conversion, [7][8][9] then stored in the memory unit. [10][11][12] The computational task is further performed by transferring the digital data between memory and a local computation unit. [13][14][15][16][17][18] The architecture of the separated sensor, memory, and data processing units results in the data-accessing latency and relatively high-power consumption.…”

Multifunctional Two-Terminal Optoelectronic Synapse Based on Zinc Oxide/Poly(3-hexylthiophene) Heterojunction for Neuromorphic Computing

“…In the past decade, a number of semiconductor materials have been adopted to fabricate artificial synaptic devices for constructing neuromorphic computing. − Meantime, different types of devices, such as memristor, atomic switch, phase change memory (PCM), and field-effect transistor (FET) have been proposed to realize some important functions of biological synapses, including pair-pulse facilitation/depression (PPF/PPD), spike-time-dependent plasticity (STDP), spike-rate-dependent plasticity (SRDP), etc. − …”