3D-integrated multilayered physical reservoir array for learning and forecasting time-series information.

A wide reservoir computing system is an advanced architecture composed of multiple reservoir layers in parallel, which enables more complex and diverse internal dynamics for multiple time-series information processing. However, its hardware implementation has not yet been realized due to the lack of a high-performance physical reservoir and the complexity of fabricating multiple stacks. Here, we achieve a proof-of-principle demonstration of such hardware made of a multilayered three-dimensional stacked 3 × 10 × 10 tungsten oxide memristive crossbar array, with which we further realize a wide physical reservoir computing for efficient learning and forecasting of multiple time-series data. Because a three-layer structure allows the seamless and effective extraction of intricate three-dimensional local features produced by various temporal inputs, it can readily outperform two-dimensional based approaches extensively studied previously. Our demonstration paves the way for wide physical reservoir computing systems capable of efficiently processing multiple dynamic time-series information.

Hardware implementation of memristor-based artificial neural networks.

Artificial Intelligence (AI) is currently experiencing a bloom driven by deep learning (DL) techniques, which rely on networks of connected simple computing units operating in parallel. The low communication bandwidth between memory and processing units in conventional von Neumann machines does not support the requirements of emerging applications that rely extensively on large sets of data. More recent computing paradigms, such as high parallelization and near-memory computing, help alleviate the data communication bottleneck to some extent, but paradigm- shifting concepts are required. Memristors, a novel beyond-complementary metal-oxide-semiconductor (CMOS) technology, are a promising choice for memory devices due to their unique intrinsic device-level properties, enabling both storing and computing with a small, massively-parallel footprint at low power. Theoretically, this directly translates to a major boost in energy efficiency and computational throughput, but various practical challenges remain. In this work we review the latest efforts for achieving hardware-based memristive artificial neural networks (ANNs), describing with detail the working principia of each block and the different design alternatives with their own advantages and disadvantages, as well as the tools required for accurate estimation of performance metrics. Ultimately, we aim to provide a comprehensive protocol of the materials and methods involved in memristive neural networks to those aiming to start working in this field and the experts looking for a holistic approach.

High-Reliability and Self-Rectifying Alkali Ion Memristor through Bottom Electrode Design and Dopant Incorporation.

Ionic memristor devices are crucial for efficient artificial neural network computations in neuromorphic hardware. They excel in multi-bit implementation but face challenges like device reliability and sneak currents in crossbar array architecture (CAA). Interface-type ionic memristors offer low variation, self-rectification, and no forming process, making them suitable for CAA. However, they suffer from slow weight updates and poor retention and endurance. To address these issues, the study demonstrated an alkali ion self-rectifying memristor with an alkali metal reservoir formed by a bottom electrode design. By adopting Li metal as the adhesion layer of the bottom electrode, an alkali ion reservoir was formed at the bottom of the memristor layer by diffusion occurring during the atomic layer deposition process for the Na:TiO2 memristor layer. In addition, Al dopant was used to improve the retention characteristics by suppressing the diffusion of alkali cations. In the memristor device with optimized Al doping, retention characteristics of more than 20 h at 125 °C, endurance characteristics of more than 5.5 × 105, and high linearity/symmetry of weight update characteristics were achieved. In reliability tests on 100 randomly selected devices from a 32 × 32 CAA device, device-to-device and cycle-to-cycle variations showed low variation values within 81% and 8%, respectively.

Programming memristor arrays with arbitrarily high precision for analog computing.

In-memory computing represents an effective method for modeling complex physical systems that are typically challenging for conventional computing architectures but has been hindered by issues such as reading noise and writing variability that restrict scalability, accuracy, and precision in high-performance computations. We propose and demonstrate a circuit architecture and programming protocol that converts the analog computing result to digital at the last step and enables low-precision analog devices to perform high-precision computing. We use a weighted sum of multiple devices to represent one number, in which subsequently programmed devices are used to compensate for preceding programming errors. With a memristor system-on-chip, we experimentally demonstrate high-precision solutions for multiple scientific computing tasks while maintaining a substantial power efficiency advantage over conventional digital approaches.

Halide Perovskites-Based Diffusive Memristors for Artificial Mechano-Nociceptive System.

Numerous efforts for emulating organ systems comprised of multiple functional units have driven substantial advancements in bio-realistic electronics and systems. The resistance change behavior observed in diffusive memristors shares similarities with the potential change in biological neurons. Here, the diffusive threshold switching phenomenon in Ag-incorporated organometallic halide perovskites is utilized to demonstrate the functions of afferent neurons. Halide perovskites-based diffusive memristors show a low threshold voltage of ≈0.2 V with little variation, attributed to the facile migration of Ag ions uniformly dispersed within the halide matrix. Based on the reversible and reliable volatile threshold switching, the memristors successfully demonstrate fundamental nociceptive functions including threshold firing, relaxation, and sensitization. Furthermore, to replicate the biological mechano-nociceptive phenomenon at a system level, an artificial mechano-nociceptive system is built by integrating a diffusive memristor with a force-sensing resistor. The presented system is capable of detecting and discerning the detrimental impact caused by a heavy steel ball, effectively exhibiting the corresponding sensitization response. By further extending the single nociceptive system into a 5 × 5 array, successful stereoscopic nociception of uneven impulses is achieved in the artificial skin system through array-scale sensitization. These results represent significant progress in the field of bio-inspired electronics and systems.

Filament-free memristors for computing.

Memristors have attracted increasing attention due to their tremendous potential to accelerate data-centric computing systems. The dynamic reconfiguration of memristive devices in response to external electrical stimuli can provide highly desirable novel functionalities for computing applications when compared with conventional complementary-metal-oxide-semiconductor (CMOS)-based devices. Those most intensively studied and extensively reviewed memristors in the literature so far have been filamentary type memristors, which typically exhibit a relatively large variability from device to device and from switching cycle to cycle. On the other hand, filament-free switching memristors have shown a better uniformity and attractive dynamical properties, which can enable a variety of new computing paradigms but have rarely been reviewed. In this article, a wide range of filament-free switching memristors and their corresponding computing applications are reviewed. Various junction structures, switching properties, and switching principles of filament-free memristors are surveyed and discussed. Furthermore, we introduce recent advances in different computing schemes and their demonstrations based on non-filamentary memristors. This Review aims to present valuable insights and guidelines regarding the key computational primitives and implementations enabled by these filament-free switching memristors.

Recent Advances and Future Prospects for Memristive Materials, Devices, and Systems.

Memristive technology has been rapidly emerging as a potential alternative to traditional CMOS technology, which is facing fundamental limitations in its development. Since oxide-based resistive switches were demonstrated as memristors in 2008, memristive devices have garnered significant attention due to their biomimetic memory properties, which promise to significantly improve power consumption in computing applications. Here, we provide a comprehensive overview of recent advances in memristive technology, including memristive devices, theory, algorithms, architectures, and systems. In addition, we discuss research directions for various applications of memristive technology including hardware accelerators for artificial intelligence, in-sensor computing, and probabilistic computing. Finally, we provide a forward-looking perspective on the future of memristive technology, outlining the challenges and opportunities for further research and innovation in this field. By providing an up-to-date overview of the state-of-the-art in memristive technology, this review aims to inform and inspire further research in this field.

Thousands of conductance levels in memristors integrated on CMOS.

Neural networks based on memristive devices1-3 have the ability to improve throughput and energy efficiency for machine learning4,5 and artificial intelligence6, especially in edge applications7-21. Because training a neural network model from scratch is costly in terms of hardware resources, time and energy, it is impractical to do it individually on billions of memristive neural networks distributed at the edge. A practical approach would be to download the synaptic weights obtained from the cloud training and program them directly into memristors for the commercialization of edge applications. Some post-tuning in memristor conductance could be done afterwards or during applications to adapt to specific situations. Therefore, in neural network applications, memristors require high-precision programmability to guarantee uniform and accurate performance across a large number of memristive networks22-28. This requires many distinguishable conductance levels on each memristive device, not only laboratory-made devices but also devices fabricated in factories. Analog memristors with many conductance states also benefit other applications, such as neural network training, scientific computing and even 'mortal computing'25,29,30. Here we report 2,048 conductance levels achieved with memristors in fully integrated chips with 256 × 256 memristor arrays monolithically integrated on complementary metal-oxide-semiconductor (CMOS) circuits in a commercial foundry. We have identified the underlying physics that previously limited the number of conductance levels that could be achieved in memristors and developed electrical operation protocols to avoid such limitations. These results provide insights into the fundamental understanding of the microscopic picture of memristive switching as well as approaches to enable high-precision memristors for various applications. Fig. 1 HIGH-PRECISION MEMRISTOR FOR NEUROMORPHIC COMPUTING.: a, Proposed scheme of the large-scale application of memristive neural networks for edge computing. Neural network training is performed in the cloud. The obtained weights are downloaded and accurately programmed into a massive number of memristor arrays distributed at the edge, which imposes high-precision requirements on memristive devices. b, An eight-inch wafer with memristors fabricated by a commercial semiconductor manufacturer. c, High-resolution transmission electron microscopy image of the cross-section view of a memristor. Pt and Ta serve as the bottom electrode (BE) and top electrode (TE), respectively. Scale bars, 1 µm and 100 nm (inset). d, Magnification of the memristor material stack. Scale bar, 5 nm. e, As-programmed (blue) and after-denoising (red) currents of a memristor are read by a constant voltage (0.2 V). The denoising process eliminated the large-amplitude RTN observed in the as-programmed state (see Methods). f, Magnification of three nearest-neighbour states after denoising. The current of each state was read by a constant voltage (0.2 V). No large-amplitude RTN was observed, and all of the states can be clearly distinguished. g, An individual memristor on the chip was tuned into 2,048 resistance levels by high-resolution off-chip driving circuitry, and each resistance level was read by a d.c. voltage sweeping from 0 to 0.2 V. The target resistance was set from 50 µS to 4,144 µS with a 2-µS interval between neighbouring levels. All readings at 0.2 V are less than 1 µS from the target conductance. Bottom inset, magnification of the resistance levels. Top inset, experimental results of an entire 256 × 256 array programmed by its 6-bit on-chip circuitry into 64 32 × 32 blocks, and each block is programmed into one of the 64 conductance levels. Each of the 256 × 256 memristors has been previously switched over one million cycles, demonstrating the high endurance and robustness of the devices.

Memristive Field-Programmable Analog Arrays for Analog Computing.

The increasing interests in analog computing nowadays call for multipurpose analog computing platforms with reconfigurability. The advancement of analog computing, enabled by novel electronic elements like memristors, has shown its potential to sustain the exponential growth of computing demand in the new era of analog data deluge. Here, a platform of a memristive field-programmable analog array (memFPAA) is experimentally demonstrated with memristive devices serving as a variety of core analog elements and CMOS components as peripheral circuits. The memFPAA is reconfigured to implement a first-order band pass filter, an audio equalizer, and an acoustic mixed frequency classifier, as application examples. The memFPAA, featured with programmable analog memristors, memristive routing networks, and memristive vector-matrix multipliers, opens opportunities for fast prototyping analog designs as well as efficient analog applications in signal processing and neuromorphic computing.

Standards for the Characterization of Endurance in Resistive Switching Devices.

Resistive switching (RS) devices are emerging electronic components that could have applications in multiple types of integrated circuits, including electronic memories, true random number generators, radiofrequency switches, neuromorphic vision sensors, and artificial neural networks. The main factor hindering the massive employment of RS devices in commercial circuits is related to variability and reliability issues, which are usually evaluated through switching endurance tests. However, we note that most studies that claimed high endurances >106 cycles were based on resistance versus cycle plots that contain very few data points (in many cases even <20), and which are collected in only one device. We recommend not to use such a characterization method because it is highly inaccurate and unreliable (i.e., it cannot reliably demonstrate that the device effectively switches in every cycle and it ignores cycle-to-cycle and device-to-device variability). This has created a blurry vision of the real performance of RS devices and in many cases has exaggerated their potential. This article proposes and describes a method for the correct characterization of switching endurance in RS devices; this method aims to construct endurance plots showing one data point per cycle and resistive state and combine data from multiple devices. Adopting this recommended method should result in more reliable literature in the field of RS technologies, which should accelerate their integration in commercial products.

A fully hardware-based memristive multilayer neural network.

Memristive crossbar arrays promise substantial improvements in computing throughput and power efficiency through in-memory analog computing. Previous machine learning demonstrations with memristive arrays, however, relied on software or digital processors to implement some critical functionalities, leading to frequent analog/digital conversions and more complicated hardware that compromises the energy efficiency and computing parallelism. Here, we show that, by implementing the activation function of a neural network in analog hardware, analog signals can be transmitted to the next layer without unnecessary digital conversion, communication, and processing. We have designed and built compact rectified linear units, with which we constructed a two-layer perceptron using memristive crossbar arrays, and demonstrated a recognition accuracy of 93.63% for the Modified National Institute of Standard and Technology (MNIST) handwritten digits dataset. The fully hardware-based neural network reduces both the data shuttling and conversion, capable of delivering much higher computing throughput and power efficiency.

Roadmap on emerging hardware and technology for machine learning.

Recent progress in artificial intelligence is largely attributed to the rapid development of machine learning, especially in the algorithm and neural network models. However, it is the performance of the hardware, in particular the energy efficiency of a computing system that sets the fundamental limit of the capability of machine learning. Data-centric computing requires a revolution in hardware systems, since traditional digital computers based on transistors and the von Neumann architecture were not purposely designed for neuromorphic computing. A hardware platform based on emerging devices and new architecture is the hope for future computing with dramatically improved throughput and energy efficiency. Building such a system, nevertheless, faces a number of challenges, ranging from materials selection, device optimization, circuit fabrication and system integration, to name a few. The aim of this Roadmap is to present a snapshot of emerging hardware technologies that are potentially beneficial for machine learning, providing the Nanotechnology readers with a perspective of challenges and opportunities in this burgeoning field.

Integration and Co-design of Memristive Devices and Algorithms for Artificial Intelligence.

Memristive devices share remarkable similarities to biological synapses, dendrites, and neurons at both the physical mechanism level and unit functionality level, making the memristive approach to neuromorphic computing a promising technology for future artificial intelligence. However, these similarities do not directly transfer to the success of efficient computation without device and algorithm co-designs and optimizations. Contemporary deep learning algorithms demand the memristive artificial synapses to ideally possess analog weighting and linear weight-update behavior, requiring substantial device-level and circuit-level optimization. Such co-design and optimization have been the main focus of memristive neuromorphic engineering, which often abandons the "non-ideal" behaviors of memristive devices, although many of them resemble what have been observed in biological components. Novel brain-inspired algorithms are being proposed to utilize such behaviors as unique features to further enhance the efficiency and intelligence of neuromorphic computing, which calls for collaborations among electrical engineers, computing scientists, and neuroscientists.

Power-efficient neural network with artificial dendrites.

In the nervous system, dendrites, branches of neurons that transmit signals between synapses and soma, play a critical role in processing functions, such as nonlinear integration of postsynaptic signals. The lack of these critical functions in artificial neural networks compromises their performance, for example in terms of flexibility, energy efficiency and the ability to handle complex tasks. Here, by developing artificial dendrites, we experimentally demonstrate a complete neural network fully integrated with synapses, dendrites and soma, implemented using scalable memristor devices. We perform a digit recognition task and simulate a multilayer network using experimentally derived device characteristics. The power consumption is more than three orders of magnitude lower than that of a central processing unit and 70 times lower than that of a typical application-specific integrated circuit chip. This network, equipped with functional dendrites, shows the potential of substantial overall performance improvement, for example by extracting critical information from a noisy background with significantly reduced power consumption and enhanced accuracy.

Gate-tunable van der Waals heterostructure for reconfigurable neural network vision sensor.

Early processing of visual information takes place in the human retina. Mimicking neurobiological structures and functionalities of the retina provides a promising pathway to achieving vision sensor with highly efficient image processing. Here, we demonstrate a prototype vision sensor that operates via the gate-tunable positive and negative photoresponses of the van der Waals (vdW) vertical heterostructures. The sensor emulates not only the neurobiological functionalities of bipolar cells and photoreceptors but also the unique connectivity between bipolar cells and photoreceptors. By tuning gate voltage for each pixel, we achieve reconfigurable vision sensor for simultaneous image sensing and processing. Furthermore, our prototype vision sensor itself can be trained to classify the input images by updating the gate voltages applied individually to each pixel in the sensor. Our work indicates that vdW vertical heterostructures offer a promising platform for the development of neural network vision sensor.

A Low-Current and Analog Memristor with Ru as Mobile Species.

The switching parameters and device performance of memristors are predominately determined by their mobile species and matrix materials. Devices with oxygen or oxygen vacancies as the mobile species usually exhibit a great retention but also need a relatively high switching current (e.g., >30 µA), while devices with Ag or Cu as cation mobile species do not require a high switching current but usually show a poor retention. Here, Ru is studied as a new type of mobile species for memristors to achieve low switching current, fast speed, good reliability, scalability, and analog switching property simultaneously. An electrochemical metallization-like memristor with a stack of Pt/Ta2 O5 /Ru is developed. Migration of Ru ions is revealed by energy-dispersive X-ray spectroscopy mapping and in situ transmission electron microscopy within a sub-10 nm active device area before and after switching. The results open up a new avenue to engineer memristors for desired properties.

An artificial spiking afferent nerve based on Mott memristors for neurorobotics.

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 NbOx 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.

Fully hardware-implemented memristor convolutional neural network.

Memristor-enabled neuromorphic computing systems provide a fast and energy-efficient approach to training neural networks1-4. However, convolutional neural networks (CNNs)-one of the most important models for image recognition5-have not yet been fully hardware-implemented using memristor crossbars, which are cross-point arrays with a memristor device at each intersection. Moreover, achieving software-comparable results is highly challenging owing to the poor yield, large variation and other non-ideal characteristics of devices6-9. Here we report the fabrication of high-yield, high-performance and uniform memristor crossbar arrays for the implementation of CNNs, which integrate eight 2,048-cell memristor arrays to improve parallel-computing efficiency. In addition, we propose an effective hybrid-training method to adapt to device imperfections and improve the overall system performance. We built a five-layer memristor-based CNN to perform MNIST10 image recognition, and achieved a high accuracy of more than 96 per cent. In addition to parallel convolutions using different kernels with shared inputs, replication of multiple identical kernels in memristor arrays was demonstrated for processing different inputs in parallel. The memristor-based CNN neuromorphic system has an energy efficiency more than two orders of magnitude greater than that of state-of-the-art graphics-processing units, and is shown to be scalable to larger networks, such as residual neural networks. Our results are expected to enable a viable memristor-based non-von Neumann hardware solution for deep neural networks and edge computing.