Neuromorphic one-shot learning utilizing a phase-transition material.

Design of hardware based on biological principles of neuronal computation and plasticity in the brain is a leading approach to realizing energy- and sample-efficient AI and learning machines. An important factor in selection of the hardware building blocks is the identification of candidate materials with physical properties suitable to emulate the large dynamic ranges and varied timescales of neuronal signaling. Previous work has shown that the all-or-none spiking behavior of neurons can be mimicked by threshold switches utilizing material phase transitions. Here, we demonstrate that devices based on a prototypical metal-insulator-transition material, vanadium dioxide (VO2), can be dynamically controlled to access a continuum of intermediate resistance states. Furthermore, the timescale of their intrinsic relaxation can be configured to match a range of biologically relevant timescales from milliseconds to seconds. We exploit these device properties to emulate three aspects of neuronal analog computation: fast (~1 ms) spiking in a neuronal soma compartment, slow (~100 ms) spiking in a dendritic compartment, and ultraslow (~1 s) biochemical signaling involved in temporal credit assignment for a recently discovered biological mechanism of one-shot learning. Simulations show that an artificial neural network using properties of VO2 devices to control an agent navigating a spatial environment can learn an efficient path to a reward in up to fourfold fewer trials than standard methods. The phase relaxations described in our study may be engineered in a variety of materials and can be controlled by thermal, electrical, or optical stimuli, suggesting further opportunities to emulate biological learning in neuromorphic hardware.

Bioinspired iontronic synapse fibers for ultralow-power multiplexing neuromorphic sensorimotor textiles.

Artificial neuromorphic devices can emulate dendric integration, axonal parallel transmission, along with superior energy efficiency in facilitating efficient information processing, offering enormous potential for wearable electronics. However, integrating such circuits into textiles to achieve biomimetic information perception, processing, and control motion feedback remains a formidable challenge. Here, we engineer a quasi-solid-state iontronic synapse fiber (ISF) comprising photoresponsive TiO2, ion storage Co-MoS2, and an ion transport layer. The resulting ISF achieves inherent short-term synaptic plasticity, femtojoule-range energy consumption, and the ability to transduce chemical/optical signals. Multiple ISFs are interwoven into a synthetic neural fabric, allowing the simultaneous propagation of distinct optical signals for transmitting parallel information. Importantly, IFSs with multiple input electrodes exhibit spatiotemporal information integration. As a proof of concept, a textile-based multiplexing neuromorphic sensorimotor system is constructed to connect synaptic fibers with artificial fiber muscles, enabling preneuronal sensing information integration, parallel transmission, and postneuronal information output to control the coordinated motor of fiber muscles. The proposed fiber system holds enormous promise in wearable electronics, soft robotics, and biomedical engineering.

Machine learning without a processor: Emergent learning in a nonlinear analog network.

Standard deep learning algorithms require differentiating large nonlinear networks, a process that is slow and power-hungry. Electronic contrastive local learning networks (CLLNs) offer potentially fast, efficient, and fault-tolerant hardware for analog machine learning, but existing implementations are linear, severely limiting their capabilities. These systems differ significantly from artificial neural networks as well as the brain, so the feasibility and utility of incorporating nonlinear elements have not been explored. Here, we introduce a nonlinear CLLN-an analog electronic network made of self-adjusting nonlinear resistive elements based on transistors. We demonstrate that the system learns tasks unachievable in linear systems, including XOR (exclusive or) and nonlinear regression, without a computer. We find our decentralized system reduces modes of training error in order (mean, slope, curvature), similar to spectral bias in artificial neural networks. The circuitry is robust to damage, retrainable in seconds, and performs learned tasks in microseconds while dissipating only picojoules of energy across each transistor. This suggests enormous potential for fast, low-power computing in edge systems like sensors, robotic controllers, and medical devices, as well as manufacturability at scale for performing and studying emergent learning.

Iontronic analog of synaptic plasticity: Hydrogel-based ionic diode with chemical precipitation and dissolution.

In this study, an aqueous nonlinear synaptic element showing plasticity behavior is developed, which is based on the chemical processes in an ionic diode. The device is simple, fully ionic, and easily configurable, requiring only two terminals-for input and output-similar to biological synapses. The key processes realizing the plasticity features are chemical precipitation and dissolution, which occur at forward- or reverse-biased ionic diode junctions in appropriate reservoir electrolytes. Given that the precipitate acts as a physical barrier in the circuit, the above processes change the diode conductivity, which can be interpreted as adjusting "synaptic weight" of the system. By varying the operating conditions, we first demonstrate the four types of plasticity that can be found in biological system: long-term potentiation/depression and short-term potentiation/depression. The plasticity of the proposed iontronic device has characteristics similar to those of neural synapses. To demonstrate its potential use in comparatively complex information processing, we develop a precipitation-based iontronic synapse (PIS) capable of both potentiation and depression. Finally, we show that the postsynaptic signals from the multiple excitatory or inhibitory PISs can be integrated into the total "dendritic" current, which is a function of time and input history, as in actual hippocampal neural circuits.

Artificial neuromorphic cognitive skins based on distributed biaxially stretchable elastomeric synaptic transistors.

Cephalopod (e.g., squid, octopus, etc.) skin is a soft cognitive organ capable of elastic deformation, visualizing, stealth, and camouflaging through complex biological processes of sensing, recognition, neurologic processing, and actuation in a noncentralized, distributed manner. However, none of the existing artificial skin devices have shown distributed neuromorphic processing and cognition capabilities similar to those of a cephalopod skin. Thus, the creation of an elastic, biaxially stretchy device with embedded, distributed neurologic and cognitive functions mimicking a cephalopod skin can play a pivotal role in emerging robotics, wearables, skin prosthetics, bioelectronics, etc. This paper introduces artificial neuromorphic cognitive skins based on arrayed, biaxially stretchable synaptic transistors constructed entirely out of elastomeric materials. Systematic investigation of the synaptic characteristics such as the excitatory postsynaptic current, paired-pulse facilitation index of the biaxially stretchable synaptic transistor under various levels of biaxial mechanical strain sets the operational foundation for stretchy distributed synapse arrays and neuromorphic cognitive skin devices. The biaxially stretchy arrays here achieved neuromorphic cognitive functions, including image memorization, long-term memorization, fault tolerance, programming, and erasing functions under 30% biaxial mechanical strain. The stretchy neuromorphic imaging sensory skin devices showed stable neuromorphic pattern reinforcement performance under both biaxial and nonuniform local deformation.

Surrogate gradients for analog neuromorphic computing.

To rapidly process temporal information at a low metabolic cost, biological neurons integrate inputs as an analog sum, but communicate with spikes, binary events in time. Analog neuromorphic hardware uses the same principles to emulate spiking neural networks with exceptional energy efficiency. However, instantiating high-performing spiking networks on such hardware remains a significant challenge due to device mismatch and the lack of efficient training algorithms. Surrogate gradient learning has emerged as a promising training strategy for spiking networks, but its applicability for analog neuromorphic systems has not been demonstrated. Here, we demonstrate surrogate gradient learning on the BrainScaleS-2 analog neuromorphic system using an in-the-loop approach. We show that learning self-corrects for device mismatch, resulting in competitive spiking network performance on both vision and speech benchmarks. Our networks display sparse spiking activity with, on average, less than one spike per hidden neuron and input, perform inference at rates of up to 85,000 frames per second, and consume less than 200 mW. In summary, our work sets several benchmarks for low-energy spiking network processing on analog neuromorphic hardware and paves the way for future on-chip learning algorithms.

Evidence for long-term potentiation in phospholipid membranes.

Biological supramolecular assemblies, such as phospholipid bilayer membranes, have been used to demonstrate signal processing via short-term synaptic plasticity (STP) in the form of paired pulse facilitation and depression, emulating the brain's efficiency and flexible cognitive capabilities. However, STP memory in lipid bilayers is volatile and cannot be stored or accessed over relevant periods of time, a key requirement for learning. Using droplet interface bilayers (DIBs) composed of lipids, water and hexadecane, and an electrical stimulation training protocol featuring repetitive sinusoidal voltage cycling, we show that DIBs displaying memcapacitive properties can also exhibit persistent synaptic plasticity in the form of long-term potentiation (LTP) associated with capacitive energy storage in the phospholipid bilayer. The time scales for the physical changes associated with the LTP range between minutes and hours, and are substantially longer than previous STP studies, where stored energy dissipated after only a few seconds. STP behavior is the result of reversible changes in bilayer area and thickness. On the other hand, LTP is the result of additional molecular and structural changes to the zwitterionic lipid headgroups and the dielectric properties of the lipid bilayer that result from the buildup of an increasingly asymmetric charge distribution at the bilayer interfaces.

Memristive Switching Mechanism in Colloidal InP/ZnSe/ZnS Quantum Dot-Based Synaptic Devices for Neuromorphic Computing.

Quantum dots (QDs) have garnered a significant amount of attention as promising memristive materials owing to their size-dependent tunable bandgap, structural stability, and high level of applicability for neuromorphic computing. Despite these advantageous properties, the development of QD-based memristors has been hindered by challenges in understanding and adjusting the resistive switching (RS) behavior of QDs. Herein, we propose three types of InP/ZnSe/ZnS QD-based memristors to elucidate the RS mechanism, employing a thin poly(methyl methacrylate) layer. This approach not only allows us to identify which carriers (electron or hole) are trapped within the QD layer but also successfully demonstrates QD-based synaptic devices. Furthermore, to utilize the QD memristor as a synapse, long-term potentiation/depression (LTP/LTD) characteristics are measured, resulting in a low nonlinearity of LTP/LTD at 0.1/1. On the basis of the LTP/LTD characteristics, single-layer perceptron simulations were performed using the Extended Modified National Institute of Standards and Technology, verifying a maximum recognition rate of 91.46%.

The Investigation of Neuromimetic Dynamics in Ferroelectrics via In Situ TEM.

The core task of neuromorphic devices is to effectively simulate the behavior of neurons and synapses. Based on the functionality of ferroelectric domains with the advantages of low power consumption and high-speed response, great progress has been made in realizing neuromimetic behaviors such as ferroelectric synaptic devices. However, the correlation between the ferroelectric domain dynamics and neuromimetic behavior remains unclear. Here, we reveal the correlation between domain/domain wall dynamics and neuromimetic behaviors from a microscopic perspective in real-time by using high temporal and spatial resolution in situ transmission electron microscopy. Furthermore, we propose utilizing ferroelectric microstructures for the simultaneous simulation of neuronal and synaptic plasticity, which is expected to improve the integration and performance of ferroelectric neuromorphic devices. We believe that this work to study neuromimetic behavior from the perspective of domain dynamics is instructive for the development of ferroelectric neuromorphic devices.

Programmable Threshold Logic Implementations in a Memristor Crossbar Array.

In this study, we demonstrate the implementation of programmable threshold logics using a 32 × 32 memristor crossbar array. Thanks to forming-free characteristics obtained by the annealing process, its accurate programming characteristics are presented by a 256-level grayscale image. By simultaneous subtraction between weighted sum and threshold values with a differential pair in an opposite way, 3-input and 4-input Boolean logics are implemented in the crossbar without additional reference bias. Also, we verify a full-adder circuit and analyze its fidelity, depending on the device programming accuracy. Lastly, we successfully implement a 4-bit ripple carry adder in the crossbar and achieve reliable operations by read-based logic operations. Compared to stateful logic driven by device switching, a 4-bit ripple carry adder on a memristor crossbar array can perform more reliably in fewer steps thanks to its read-based parallel logic operation.

Self-Rectifying All-Optical Modulated Optoelectronic Multistates Memristor Crossbar Array for Neuromorphic Computing.

Researching optoelectronic memristors capable of integrating sensory and processing functions is essential for advancing the development of efficient neuromorphic vision. Here, we experimentally demonstrated an all-optical controlled and self-rectifying optoelectronic memristor (OEM) crossbar array with the function of multilevel storage under light stimuli. The NiO/TiO2 device exhibits an ultrahigh (>104) rectifying ratio (RR) thus overcoming the presence of sneak current. The reversible conductance modulation without electric signal involvement provides a novel way to realize ultrafast information processing. The proposed OEM array realized synaptic functions observed in the human brain, including long-term potentiation (LTP), long-term depression (LTD), paired-pulse facilitation (PPF), the transition from short-term memory (STM) to long-term memory (LTM), and learning experience behaviors successfully. The authors present a novel OEM crossbar that possesses complete light-modulation capabilities, potentially advancing the future development of efficient neuromorphic vision.

Electronically Reconfigurable Memristive Neuron Capable of Operating in Both Excitation and Inhibition Modes.

Threshold switching (TS) memristors are promising candidates for artificial neurons in neuromorphic systems. However, they often lack biological plausibility, typically functioning solely in an excitation mode. The absence of an inhibitory mode limits neurons' ability to synergistically process both excitatory and inhibitory synaptic signals. To address this limitation, we propose a novel memristive neuron capable of operating in both excitation and inhibition modes. The memristor's threshold voltage can be reversibly tuned using voltages of different polarities because of its bipolar TS behavior, enabling the device to function as an electronically reconfigurable bi-mode neuron. A variety of neuronal activities such as all-or-nothing behavior and tunable firing probability are mimicked under both excitatory and inhibitory stimuli. Furthermore, we develop a self-adaptive neuromorphic vision sensor based on bi-mode neurons, demonstrating effective object recognition in varied lighting conditions. Thus, our bi-mode neuron offers a versatile platform for constructing neuromorphic systems with rich functionality.

Inorganic Halide Perovskite Nanowires/Conjugated Polymer Heterojunction-Based Optoelectronic Synaptic Transistors for Dynamic Machine Vision.

Inspired by the retina, artificial optoelectronic synapses have groundbreaking potential for machine vision. The field-effect transistor is a crucial platform for optoelectronic synapses that is highly sensitive to external stimuli and can modulate conductivity. On the basis of the decent optical absorption, perovskite materials have been widely employed for constructing optoelectronic synaptic transistors. However, the reported optoelectronic synaptic transistors focus on the static processing of independent stimuli at different moments, while the natural visual information consists of temporal signals. Here, we report CsPbBrI2 nanowire-based optoelectronic synaptic transistors to study the dynamic responses of artificial synaptic transistors to time-varying visual information for the first time. Moreover, on the basis of the dynamic synaptic behavior, a hardware system with an accuracy of 85% is built to the trajectory of moving objects. This work offers a new way to develop artificial optoelectronic synapses for the construction of dynamic machine vision systems.

Graphene-Based Artificial Dendrites for Bio-Inspired Learning in Spiking Neuromorphic Systems.

Analog neuromorphic computing systems emulate the parallelism and connectivity of the human brain, promising greater expressivity and energy efficiency compared to those of digital systems. Though many devices have emerged as candidates for artificial neurons and artificial synapses, there have been few device candidates for artificial dendrites. In this work, we report on biocompatible graphene-based artificial dendrites (GrADs) that can implement dendritic processing. By using a dual side-gate configuration, current applied through a Nafion membrane can be used to control device conductance across a trilayer graphene channel, showing spatiotemporal responses of leaky recurrent, alpha, and Gaussian dendritic potentials. The devices can be variably connected to enable higher-order neuronal responses, and we show through data-driven spiking neural network simulations that spiking activity is reduced by ≤15% without accuracy loss while low-frequency operation is stabilized. This positions the GrADs as strong candidates for energy efficient bio-interfaced spiking neural networks.

Novel Three-Dimensional Artificial Neural Network Based on an Eight-Layer Vertical Memristor with an Ultrahigh Rectify Ratio (>107) and an Ultrahigh Nonlinearity (>105) for Neuromorphic Computing.

In recent years, memristors have successfully demonstrated their significant potential in artificial neural networks (ANNs) and neuromorphic computing. Nonetheless, ANNs constructed by crossbar arrays suffer from cross-talk issues and low integration densities. Here, we propose an eight-layer three-dimensional (3D) vertical crossbar memristor with an ultrahigh rectify ratio (RR > 107) and an ultrahigh nonlinearity (>105) to overcome these limitations, which enables it to reach a >1 Tb array size without reading failure. Furthermore, the proposed 3D RRAM shows advanced endurance (>1010 cycles), retention (>104 s), and uniformity. In addition, several synaptic functions observed in the human brain were mimicked. On the basis of the advanced performance, we constructed a novel 3D ANN, whose learning efficiency and recognition accuracy were enhanced significantly compared with those of conventional single-layer ANNs. These findings hold promise for the development of highly efficient, precise, integrated, and stable VLSI neuromorphic computing systems.

Organic Synaptic Transistors with Environmentally Friendly Core/Shell Quantum Dots for Wavelength-Selective Memory and Neuromorphic Functions.

Organic transistors based on organic semiconductors together with quantum dots (QDs) are attracting more and more interest because both materials have excellent optoelectronic properties and solution processability. Electronics based on nontoxic QDs are highly desired considering the potential health risks but are limited by elevated surface defects, inadequate stability, and diminished luminescent efficiency. Herein, organic synaptic transistors based on environmentally friendly ZnSe/ZnS core/shell QDs with passivating surface defects are developed, exhibiting optically programmable and electrically erasable characteristics. The synaptic transistors feature linear multibit storage capability and wavelength-selective memory function with a retention time above 6000 s. Various neuromorphic applications, including memory enhancement, optical communication, and memory consolidation behaviors, are simulated. Utilizing an established neuromorphic model, accuracies of 92% and 91% are achieved in pattern recognition and complicated electrocardiogram signal processing, respectively. This research highlights the potential of environmentally friendly QDs in neuromorphic applications and health monitoring.

Gate-Controlled Neuromorphic Functional Transition in an Electrochemical Graphene Transistor.

Neuromorphic devices have attracted significant attention as potential building blocks for the next generation of computing technologies owing to their ability to emulate the functionalities of biological nervous systems. The essential components in artificial neural networks such as synapses and neurons are predominantly implemented by dedicated devices with specific functionalities. In this work, we present a gate-controlled transition of neuromorphic functions between artificial neurons and synapses in monolayer graphene transistors that can be employed as memtransistors or synaptic transistors as required. By harnessing the reliability of reversible electrochemical reactions between carbon atoms and hydrogen ions, we can effectively manipulate the electric conductivity of graphene transistors, resulting in a high on/off resistance ratio, a well-defined set/reset voltage, and a prolonged retention time. Overall, the on-demand switching of neuromorphic functions in a single graphene transistor provides a promising opportunity for developing adaptive neural networks for the upcoming era of artificial intelligence and machine learning.

An Antiferromagnetic Neuromorphic Memory Based on Perpendicularly Magnetized CoO.

Antiferromagnets (AFMs) are ideal materials to boost neuromorphic computing toward the ultrahigh speed and ultracompact integration regime. However, developing a suitable AFM neuromorphic memory remains an aspirational but challenging goal. In this work, we construct such a memory based on the CoO/Pt heterostructure, in which the collinear insulating AFM CoO shows a strong perpendicular anisotropy facilitating its electrical readout and writing. Utilizing the unique nonlinear response and bipolar fading memory properties of the device, we demonstrate a multidimensional reservoir computing beyond the traditional binary paradigm. These results are expected to pave the way toward next-generation fast and massive neuromorphic computing.

Novel Two-Terminal Synapse/Neuron Based on an Antiferroelectric Hafnium Zirconium Oxide Device for Neuromorphic Computing.

Functionally diverse devices with artificial neuron and synapse properties are critical for neuromorphic systems. We present a two-terminal artificial leaky-integrate-fire (LIF) neuron based on 6 nm Hf0.1Zr0.9O2 (HZO) antiferroelectric (AFE) thin films and develop a synaptic device through work function (WF) engineering. LIF neuron characteristics, including integration, firing, and leakage, are achieved in W/HZO/W devices due to the accumulated polarization and spontaneous depolarization of AFE HZO films. By engineering the top electrode with asymmetric WFs, we found that Au/Ti/HZO/W devices exhibit synaptic weight plasticity, such as paired-pulse facilitation and long-term potentiation/depression, achieving >90% accuracy in digit recognition within constructed artificial neural network systems. These findings suggest that AFE HZO capacitor-based neurons and WF-engineered artificial synapses hold promise for constructing efficient spiking neuron networks and artificial neural networks, thereby advancing neuromorphic computing applications based on emerging AFE HZO devices.

Perovskite-Doped Modulated Color-Selective Photosynaptic Transistors for Target Object Recognition.

The processing of multicolor noisy images in visual neuromorphic devices requires selective absorption at specific wavelengths; however, it is difficult to achieve this because the spectral absorption range of the device is affected by the type of material. Surprisingly, the absorption range of perovskite materials can be adjusted by doping. Herein, a CdCl2 co-doped CsPbBr3 nanocrystal-based photosensitive synaptic transistor (PST) is reported. By decreasing the doping concentration, the response of the PST to short-wavelength light is gradually enhanced, and even weak light of 40 µW·cm-2 can be detected. Benefiting from the excellent color selectivity of the PST device, the device array is applied to feature extraction of target blue items and removal of red and green noise, which results in the recognition accuracy of 95% for the noisy MNIST data set. This work provides new ideas for the application of novel transistors integrating sensors and storage computing.