Abstract
The resistive switching effect in memristors typically stems from the formation and rupture of localized conductive filament paths, and HfO
2
has been accepted as one of the most promising ...resistive switching materials. However, the dynamic changes in the resistive switching process, including the composition and structure of conductive filaments, and especially the evolution of conductive filament surroundings, remain controversial in HfO
2
-based memristors. Here, the conductive filament system in the amorphous HfO
2
-based memristors with various top electrodes is revealed to be with a quasi-core-shell structure consisting of metallic hexagonal-Hf
6
O and its crystalline surroundings (monoclinic or tetragonal HfO
x
). The phase of the HfO
x
shell varies with the oxygen reservation capability of the top electrode. According to extensive high-resolution transmission electron microscopy observations and ab initio calculations, the phase transition of the conductive filament shell between monoclinic and tetragonal HfO
2
is proposed to depend on the comprehensive effects of Joule heat from the conductive filament current and the concentration of oxygen vacancies. The quasi-core-shell conductive filament system with an intrinsic barrier, which prohibits conductive filament oxidation, ensures the extreme scalability of resistive switching memristors. This study renovates the understanding of the conductive filament evolution in HfO
2
-based memristors and provides potential inspirations to improve oxide memristors for nonvolatile storage-class memory applications.
A self-organizing map (SOM) is a powerful unsupervised learning neural network for analyzing high-dimensional data in various applications. However, hardware implementation of SOM is challenging ...because of the complexity in calculating the similarities and determining neighborhoods. We experimentally demonstrated a memristor-based SOM based on Ta/TaO
/Pt 1T1R chips for the first time, which has advantages in computing speed, throughput, and energy efficiency compared with the CMOS digital counterpart, by utilizing the topological structure of the array and physical laws for computing without complicated circuits. We employed additional rows in the crossbar arrays and identified the best matching units by directly calculating the similarities between the input vectors and the weight matrix in the hardware. Using the memristor-based SOM, we demonstrated data clustering, image processing and solved the traveling salesman problem with much-improved energy efficiency and computing throughput. The physical implementation of SOM in memristor crossbar arrays extends the capability of memristor-based neuromorphic computing systems in machine learning and artificial intelligence.
Neuromorphic computing based on spikes offers great potential in highly efficient computing paradigms. Recently, several hardware implementations of spiking neural networks based on traditional ...complementary metal-oxide semiconductor technology or memristors have been developed. However, an interface (called an afferent nerve in biology) with the environment, which converts the analog signal from sensors into spikes in spiking neural networks, is yet to be demonstrated. Here we propose and experimentally demonstrate an artificial spiking afferent nerve based on highly reliable NbO
Mott memristors for the first time. The spiking frequency of the afferent nerve is proportional to the stimuli intensity before encountering noxiously high stimuli, and then starts to reduce the spiking frequency at an inflection point. Using this afferent nerve, we further build a power-free spiking mechanoreceptor system with a passive piezoelectric device as the tactile sensor. The experimental results indicate that our afferent nerve is promising for constructing self-aware neurorobotics in the future.
Spiking neural networks (SNNs) sharing large similarity with biological nervous systems are promising to process spatiotemporal information and can provide highly time‐ and energy‐efficient ...computational paradigms for the Internet‐of‐Things and edge computing. Nonvolatile electrolyte‐gated transistors (EGTs) provide prominent analog switching performance, the most critical feature of synaptic element, and have been recently demonstrated as a promising synaptic device. However, high performance, large‐scale EGT arrays, and EGT application for spatiotemporal information processing in an SNN are yet to be demonstrated. Here, an oxide‐based EGT employing amorphous Nb2O5 and LixSiO2 is introduced as the channel and electrolyte gate materials, respectively, and integrated into a 32 × 32 EGT array. The engineered EGTs show a quasi‐linear update, good endurance (106) and retention, a high switching speed of 100 ns, ultralow readout conductance (<100 nS), and ultralow areal switching energy density (20 fJ µm−2). The prominent analog switching performance is leveraged for hardware implementation of an SNN with the capability of spatiotemporal information processing, where spike sequences with different timings are able to be efficiently learned and recognized by the EGT array. Finally, this EGT‐based spatiotemporal information processing is deployed to detect moving orientation in a tactile sensing system. These results provide an insight into oxide‐based EGT devices for energy‐efficient neuromorphic computing to support edge application.
An oxide‐based electrolyte‐gated transistor (EGT) with prominent analog switching performance is introduced and integrated into a 32 × 32 array. The EGT array is leveraged for hardware implementation of spiking neural networks to process spatiotemporal information and is further deployed to detect moving orientation in a tactile sensing system. These results provide an insight into oxide‐based EGT devices for energy‐efficient neuromorphic computing to support edge application.
A memristive neuron circuit which is capable of encoding input information via temporal and adaptive frequency methods within the same structure is proposed. Its multifunctionality allows switching ...between these modes by adjusting circuit capacitors. Furthermore, it is investigated how other parameters like input voltage and resistance influence output properties for different encoding methods. This research contributes to the development of more flexible neuromorphic computing systems.
A memristive neuron circuit which is capable of encoding input information via temporal and adaptive frequency methods within the same structure is proposed. Its multifunctionality allows switching between these modes by adjusting circuit capacitance. Furthermore, we investigate how parameters like input voltage and resistance influence output properties for different encoding methods.
Synaptic devices with bipolar analog resistive switching behavior are the building blocks for memristor-based neuromorphic computing. In this work, a fully complementary metal-oxide semiconductor ...(CMOS)-compatible, forming-free, and non-filamentary memristive device (Pd/Al₂O₃/TaO
/Ta) with bipolar analog switching behavior is reported as an artificial synapse for neuromorphic computing. Synaptic functions, including long-term potentiation/depression, paired-pulse facilitation (PPF), and spike-timing-dependent plasticity (STDP), are implemented based on this device; the switching energy is around 50 pJ per spike. Furthermore, for applications in artificial neural networks (ANN), determined target conductance states with little deviation (<1%) can be obtained with random initial states. However, the device shows non-linear conductance change characteristics, and a nearly linear conductance change behavior is obtained by optimizing the training scheme. Based on these results, the device is a promising emulator for biology synapses, which could be of great benefit to memristor-based neuromorphic computing.
Memristor crossbar‐based spiking neural networks (SNNs) face challenges caused by nonidealities associated with their hardware‐based neurons and synapses. The key nonidealities include electric‐field ...noise, conductance noise, and conductance drift. This study investigates the robustness of fully connected, convolutional, residual, and spike‐timing‐dependent plasticity‐based SNNs against hardware nonidealities using the MNIST, Fashion MNIST, and CIFAR10 datasets. In response to these challenges, a novel hybrid residual SNN (HRSNN) is proposed that incorporates a new neuron circuit and a weight‐dependent loss function. The HRSNN in a high‐intensity noise environment is evaluated using the neuromorphic DVS128 Gesture dataset. The achieved accuracy rate of 92.71% is only 2.15% lower than that of the noise‐free environment. These results demonstrate the robustness of the proposed HRSNN under high‐intensity noise conditions and present new possibilities for the advancement of neuromorphic computing in noisy environments.
This study explores challenges faced by memristor crossbar‐based spiking neural networks (SNNs) due to nonidealities. The researchers propose a hybrid residual SNN (HRSNN) that incorporates a new neuron circuit and a weight‐dependent loss function to enhance robustness. Evaluations using various datasets demonstrate the HRSNN's effectiveness in high‐intensity noise environments, showing potential for advancing neuromorphic computing in noisy conditions.
The human memory system plays an indispensable role in oblivion, learning, and memorization. Implementing a memory system within electronic devices contributes an important step toward constructing a ...neuromorphic system that emulates advanced mental functions of the human brain. Given the complex time‐tailoring requirement, integrating a human memory system into one system is a great challenge. Here, one van der Waals heterostructure with flexible time‐tailoring ability is demonstrated, which can meet the high requirement of human memory system programming. By stacking volatile and nonvolatile function layers, all basic memory types, including sensory memory, short‐term and long‐term memory are integrated into the device and the transition between these memory types are flexible. Moreover, decision‐making action and in situ result storage are also demonstrated. It is anticipated that the demonstrated time‐tailoring system will support the model of a human memory system.
van der Waals heterostructures with time‐tailoring ability are developed for human memory system programming. Retention time can be tailored from milliseconds to kiloseconds. Thus, a human memory system with rich functionalities is implemented in this cell and the system also has a capability of decision‐making and in situ data storage.
Cochleas are the basis for biology to process and recognize speech information, emulating which with electronic devices helps us construct high-efficient intelligent voice systems. Memristor provides ...novel physics for performing neuromorphic engineering beyond complementary metal-oxide-semiconductor technology. This work presents an artificial cochlea based on the shallen-key filter model configured with memristors, in which one filter emulates one channel. We first fabricate a memristor with the TiN/HfO
x
/TaO
x
/TiN structure to implement such a cochlea and demonstrate the non-volatile multilevel states through electrical operations. Then, we build the shallen-key filter circuit and experimentally demonstrate the frequency-selection function of cochlea’s five channels, whose central frequency is determined by the memristor’s resistance. To further demonstrate the feasibility of the cochlea for system applications, we use it to extract the speech signal features and then combine it with a convolutional neural network to recognize the
Free Spoken Digit Dataset
. The recognition accuracy reaches 92% with 64 channels, compatible with the traditional 64 Fourier transform transformation points of mel-frequency cepstral coefficients method with 95% recognition accuracy. This work provides a novel strategy for building cochleas, which has a great potential to conduct configurable, high-parallel, and high-efficient auditory systems for neuromorphic robots.