Convolutional Image Processing Research Articles

Heterogeneous ion-modulated 2D-WS2 heterosynaptic memtransistor for controllable synaptic modulation and energy-efficient neuromorphic computing

Heterosynaptic plasticity plays a key role in addressing the interactions between local and global neuromodulation and the implementation of more complex biological functions. However, conventional heterosynaptic devices suffer from high energy consumption, high operating voltages and a lack of multi-modal regulation, which hinder the development of next-generation neuromorphic computing. Here, we report a heterogeneous ion-modulated 2D-WS2 four-terminal heterosynaptic memtransistor (Na+@WSHMT) with typical heterosynaptic plasticity at low stimulation voltage (200 mV) and ultra-low energy consumption (150 aJ). Moreover, the potential of four-terminal heterosynaptic memtransistors is demonstrated for enhancing the digit recognition with an increased recognition accuracy from 90.5 % to 93.3 % by stimulating the heterosynaptic terminal (light). Furthermore, convolutional image processing such as feature extraction is successfully performed by a convolutional neural network based on the multi-conductance characteristic of Na+@WSHMT. The ion-modulated heterosynaptic memtransistor provide a promising strategy for energy-efficient and controllable neuromorphic computing.

Ag-doped non-imperfection-enabled uniform memristive neuromorphic device based on van der Waals indium phosphorus sulfide.

Memristors are considered promising energy-efficient artificial intelligence hardware, which can eliminate the von Neumann bottleneck by parallel in-memory computing. The common imperfection-enabled memristors are plagued with critical variability issues impeding their commercialization. Reported approaches to reduce the variability usually sacrifice other performances, e.g., small on/off ratios and high operation currents. Here, we demonstrate an unconventional Ag-doped nonimperfection diffusion channel-enabled memristor in van der Waals indium phosphorus sulfide, which can combine ultralow variabilities with desirable metrics. We achieve operation voltage, resistance, and on/off ratio variations down to 3.8, 2.3, and 6.9% at their extreme values of 0.2 V, 1011 ohms, and 108, respectively. Meanwhile, the operation current can be pushed from 1 nA to 1 pA at the scalability limit of 6 nm after Ag doping. Fourteen Boolean logic functions and convolutional image processing are successfully implemented by the memristors, manifesting the potential for logic-in-memory devices and efficient non-von Neumann accelerators.

High‐Performance Memristors Based on Few‐Layer Manganese Phosphorus Trisulfide for Neuromorphic Computing

AbstractWhile transition‐metal thiophosphate (MPX3) materials have been a subject of extensive research in recent years, experimental studies on MPX3‐based memristors are still in their early stages, with device performance being less than ideal. Here, the successful fabrication of high‐yield, high‐performance, and uniform memristors are demonstrated to possess desired characteristics for neuromorphic computing using a single‐crystalline few‐layered manganese phosphorus trisulfide (MnPS3) as a resistive switching medium. The Ti/MnPS3/Au memristor exhibits small switching voltage (<1 V), long memory retention (104 s), fast switching speed (≈20 ns), high On/Off ratio (nearly two orders of magnitude), and simultaneously achieves emulation of synaptic weight plasticity. The microscopic investigation of the structural and chemical characteristics of the few‐layer MnPS3 reveals the presence of structural defects and residual Ti throughout the stacked layer following the application of voltage, which contributes to the uniformity of switching with a low set voltage. With highly linear and symmetric analog weight updates coupled with the capability of accurate decimal arithmetic operations, a high accuracy of 95.15% in supervised learning using the MNIST handwritten recognition dataset is achieved in the artificial neural network. Furthermore, convolutional image processing can be implemented using hardware multiply‐and‐accumulate operation in an experimental memristor crossbar array.

High-Performance Neuromorphic Computing and Logic Operation Based on a Self-Assembled Vertically Aligned Nanocomposite SrTiO3:MgO Film Memristor.

Neuromorphic computing based on memristors capable of in-memory computing is promising to break the energy and efficiency bottleneck of well-known von Neumann architectures. However, unstable and nonlinear conductance updates compromise the recognition accuracy and block the integration of neural network hardware. To this end, we present a highly stable memristor with self-assembled vertically aligned nanocomposite (VAN) SrTiO3:MgO films that achieve excellent resistive switching with low set/reset voltage variability (4.7%/-5.6%) and highly linear conductivity variation (nonlinearity = 0.34) by spatially limiting the conductive channels at the vertical interfaces. Various synaptic behaviors are simulated by continuously modulating the conductance. Especially, convolutional image processing using diverse crossbar kernels is demonstrated, and the artificial neural network achieves an overwhelming recognition accuracy of up to 97.50% for handwritten digits. Even under the perturbation of Poisson noise (λ = 10), 6% Salt and Pepper noise, and 5% Gaussian noise, the high recognition accuracies are retained at 95.43%, 94.56%, and 95.97%, respectively. Importantly, the logic memory function is proven experimentally based on the nonvolatile properties. This work provides a material system and design idea to achieve high-performance neuromorphic computing and logic operation.

In-Memory Computing using Memristor Arrays with Ultrathin 2D PdSeOx /PdSe2 Heterostructure.

In-memory computing based on memristor arrays holds promise to address the speed and energy issues of the classical von Neumann computing system. However, the stochasticity of ions' transport in conventional oxide-based memristors imposes severe intrinsic variability, which compromises learning accuracy and hinders the implementation of neural network hardware accelerators. Here, these challenges are addressed using a low-voltage memristor array based on an ultrathin PdSeOx /PdSe2 heterostructure switching medium realized by a controllable ultraviolet (UV)-ozone treatment. A distinctively different ions' transport mechanism is revealed in the heterostructure that can confine the formation of conductive filaments, leading to a remarkable uniform switching with low set and reset voltage variability values of 4.8% and -3.6%, respectively. Moreover, convolutional image processing is further implemented using various crossbar kernels that achieve a high recognition accuracy of ≈93.4% due to the highly linear and symmetric analog weight update as well as multiple conductance states, manifesting its potential beyond von Neumann computing.