Recent progress on two-dimensional neuromorphic devices and artificial neural network

Mimicking biological synapses with microelectronic devices is widely considered as the first step in hardware building artificial neuromorphic networks, which is also the basis of brain-inspired neuromorphic computing. Numerous artificial neurons and synapses making up an artificial neuromorphic network have been gained wide attention due to their powerful and efficient data processing capabilities. Recently, artificial synapses, especially memristor-type and transistor-type synapses based on multifarious two-dimensional (2D) materials have been paid much attention. The unique properties of 2D materials make devices perform well in learning ability and power efficiency when mimicking synaptic behaviors, which highlights the feasibility of 2D neuromorphic devices in constructing artificial neuromorphic networks. Herein, the basic structures and principles of biological synapses are introduced, and the definitions of synaptic behaviors in synaptic electronic devices are discussed. Then, the progress of 2D memristor-type and transistor-type neuromorphic devices involving their device architecture, neuromorphic operational mechanism, and promising applications is reviewed. Finally, the future challenges of artificial synaptic devices based on 2D materials are discussed briefly.     

An analytical approach to neuronal connectivity

This paper describes how to analytically characterize the connectivity of neuromorphic networks taking into account the morphology of their elements. By assuming that all neurons have the same shape and are regularly distributed along a two-dimensional orthogonal lattice with parameter , we obtain the exact number of connections and cycles of any length by applying convolutions and the respective spectral density derived from the adjacency matrix. It is shown that neuronal shape plays an important role in defining the spatial distribution of synapses in neuronal networks. In addition, we observe that neuromorphic networks typically present an interesting property where the pattern of connections is progressively shifted along the spatial domain for increasing connection lengths. This arises from the fact that the axon reference point usually does not coincide with the cell center of mass of neurons. Morphological measurements for characterization of the spatial distribution of connections, including the adjacency matrix spectral density and the lacunarity of the connections, are suggested and illustrated. We also show that Hopfield networks with connectivity defined by different neuronal morphologies, which are quantified by the analytical approach proposed herein, lead to distinct performances for associative recall, as measured by the overlap index. The potential of our approach is illustrated for digital images of real neuronal cells.     

III–V quantum devices and circuits based on nanoscale Schottky gate control of hexagonal quantum wire networks

The concept, the present status, key issues and future prospects of a novel hexagonal binary decision diagram (BDD) quantum circuit approach for III–V quantum large-scale integrated circuits (QLSIs) are presented and discussed. In this approach, the BDD logic circuits are implemented on III–V semiconductor-based hexagonal nanowire networks controlled by nanoscale Schottky gates. The hexagonal BDD QLSIs can operate at delay-power products near the quantum limit in the quantum regime as well as in the many-electron classical regime. To demonstrate the feasibility of the present approach, GaAs Schottky wrap gate (WPG)-based single-electron BDD node devices and their integrated circuits were fabricated and their proper operations were confirmed. Selectively grown InGaAs sub-10 nm quantum wires and their hexagonal networks have been investigated to form high-density hexagonal BDD QLSIs operating in the quantum regime at room temperature.     

The fabrication technique and electrical properties of a free-standing GaN nanowire

We fabricated a free-standing structure of a GaN nanowire by selectively etching Si₃N₄, previously grown on a SiO₂ substrate, for application to three-dimensional integrated circuits such as nanorelays and actuators. In the nanowire-deposition process we adopted electrophoresis and reactive ion etching techniques to achieve a well-aligned and free-standing nanowire. The electrical transport measurements were performed from room temperature down to liquid-nitrogen temperature. The current–voltage (I–V) characteristics showed a rectifying behavior in the whole temperature range. We analyze this property as a Schottky barrier formation between the nanowire and electrodes. PACS  61.46.+w; 73.22.-f; 73.40.Ei; 81.07.Bc; 81.16.Rf     

Theoretical study of the migration of the hydrogen atom adsorbed on aluminum nanowire

We study the behavior of a hydrogen atom adsorbed on aluminum nanowire based on density functional theory. In this study, we focus on the electronic structure, potential energy surface (PES), and quantum mechanical effects on hydrogen and deuterium atoms. The activation energy of the diffusion of a hydrogen atom to the axis direction is derived as 0.19 eV from PES calculations. The probability density, which is calculated by including quantum effects, is localized on an aluminum top site in both cases of hydrogen and deuterium atoms of the ground state. In addition, some excited states are distributed between aluminum atoms on the surface of the nanowire. The energy difference between the ground state and these excited states are below 0.1 eV, which is much smaller than the activation energy of PES calculations. Thus using these excited states, hydrogen and deuterium atoms may move to the axial direction easily. We also discuss the electronic structure of the nanowire surface using quantum energy density defined by one of the authors.     

Identification of functional information subgraphs in complex networks

We present a general information theoretic approach for identifying functional subgraphs in complex networks. We show that the uncertainty in a variable can be written as a sum of information quantities, where each term is generated by successively conditioning mutual informations on new measured variables in a way analogous to a discrete differential calculus. The analogy to a Taylor series suggests efficient optimization algorithms for determining the state of a target variable in terms of functional groups of other nodes. We apply this methodology to electrophysiological recordings of cortical neuronal networks grown in vitro. Each cell's firing is generally explained by the activity of a few neurons. We identify these neuronal subgraphs in terms of their redundant or synergetic character and reconstruct neuronal circuits that account for the state of target cells.     

Revealing the Dynamics of Neural Information Processing with Multivariate Information Decomposition

The varied cognitive abilities and rich adaptive behaviors enabled by the animal nervous system are often described in terms of information processing. This framing raises the issue of how biological neural circuits actually process information, and some of the most fundamental outstanding questions in neuroscience center on understanding the mechanisms of neural information processing. Classical information theory has long been understood to be a natural framework within which information processing can be understood, and recent advances in the field of multivariate information theory offer new insights into the structure of computation in complex systems. In this review, we provide an introduction to the conceptual and practical issues associated with using multivariate information theory to analyze information processing in neural circuits, as well as discussing recent empirical work in this vein. Specifically, we provide an accessible introduction to the partial information decomposition (PID) framework. PID reveals redundant, unique, and synergistic modes by which neurons integrate information from multiple sources. We focus particularly on the synergistic mode, which quantifies the “higher-order” information carried in the patterns of multiple inputs and is not reducible to input from any single source. Recent work in a variety of model systems has revealed that synergistic dynamics are ubiquitous in neural circuitry and show reliable structure–function relationships, emerging disproportionately in neuronal rich clubs, downstream of recurrent connectivity, and in the convergence of correlated activity. We draw on the existing literature on higher-order information dynamics in neuronal networks to illustrate the insights that have been gained by taking an information decomposition perspective on neural activity. Finally, we briefly discuss future promising directions for information decomposition approaches to neuroscience, such as work on behaving animals, multi-target generalizations of PID, and time-resolved local analyses.     

Semi-passivity and synchronization of diffusively coupled neuronal oscillators

We discuss synchronization in networks of neuronal oscillators which are interconnected via diffusive coupling, i.e. linearly coupled via gap junctions. In particular, we present sufficient conditions for synchronization in these networks using the theory of semi-passive and passive systems. We show that the conductance based neuronal models of Hodgkin-Huxley, Morris-Lecar, and the popular reduced models of FitzHugh-Nagumo and Hindmarsh-Rose all satisfy a semi-passivity property, i.e. that is the state trajectories of such a model remain oscillatory but bounded provided that the supplied (electrical) energy is bounded. As a result, for a wide range of coupling configurations, networks of these oscillators are guaranteed to possess ultimately bounded solutions. Moreover, we demonstrate that when the coupling is strong enough the oscillators become synchronized. Our theoretical conclusions are confirmed by computer simulations with coupled Hindmarsh-Rose and Morris-Lecar oscillators. Finally we discuss possible “instabilities” in networks of oscillators induced by the diffusive coupling.     

The two independent equations of circuits in integral form of field theory:The fundamental law of circuits

1 Introduction Historically, circuit theory was initially considered as a part of the electromagnetic theory. Later on, it branched out to become an independent theory. After several stages of its development, Kirchhoff’s law was commonly regarded as the fundamental law of circuits[1]. Especially after the 1960s, the completely topological formulation of Kirchhoff’s law made even more important contribution to the development of moderncircuit theory. However, it has been also known for a l…     

Imaging the neural circuitry and chemical control of aggressive motivation

Background  

With the advent of functional magnetic resonance imaging (fMRI) in awake animals it is possible to resolve patterns of neuronal activity across the entire brain with high spatial and temporal resolution. Synchronized changes in neuronal activity across multiple brain areas can be viewed as functional neuroanatomical circuits coordinating the thoughts, memories and emotions for particular behaviors. To this end, fMRI in conscious rats combined with 3D computational analysis was used to identifying the putative distributed neural circuit involved in aggressive motivation and how this circuit is affected by drugs that block aggressive behavior.     

Learning by message passing in networks of discrete synapses

We show that a message-passing process allows us to store in binary "material" synapses a number of random patterns which almost saturate the information theoretic bounds. We apply the learning algorithm to networks characterized by a wide range of different connection topologies and of size comparable with that of biological systems (e.g., [EQUATION: SEE TEXT]). The algorithm can be turned into an online-fault tolerant-learning protocol of potential interest in modeling aspects of synaptic plasticity and in building neuromorphic devices.     

Stochastic induced dynamics in neuromorphic optoelectronic oscillators

We investigate the dynamics of optoelectronic oscillator (OEO) systems based on resonant tunneling diode photodetector (RTD-PD) and laser diode hybrid integrated circuits. We demonstrate that RTD-based OEOs can be noise-activated in either monostable or bistable operating conditions, providing a rich variety of signal outputs—spiking, square pulses, bursting—and behaviours—stochastic and coherence resonances—that are similar to that of biological systems such as neurons. The potential for fully monolithic integration of our OEO confers them a great potential in novel neuromorphic optoelectronic circuits for signal processing tasks including re-timing and re-shaping of pulsed signals exploiting either the monostable or the bistable operating conditions.     

N型局部有源忆阻器的神经形态行为

局部有源忆阻器(locally-active memristor,LAM)凭借其高集成度、低功耗和局部有源特性等优点,在神经形态计算领域显示出巨大的潜力.本文提出了一种简单的N型LAM数学模型,通过揭示其非线性动力特性,设计了N型LAM神经元电路.采用Hopf分岔、数值分析等方法定量研究了该电路的动力学行为,成功模拟了多种神经形态行为,包括全或无行为、尖峰、簇发、周期振荡等.并利用该神经元电路结构模拟了生物触觉神经元的频率特性.仿真结果表明:当输入信号幅值低于阈值时,神经元电路输出信号的振荡频率与输入信号强度呈正相关(即兴奋状态),并在阈值处达到最大值.随后,继续增大激励强度,振荡频率则逐渐降低(即保护性抑制状态).最后,设计了N型LAM硬件仿真器,并完成了人工神经元电路的硬件实现,实验结果与仿真结果、理论分析相一致,验证了该N型LAM具备的神经形态行为.     

Nanowire as pico-gram balance at workplace atmosphere

The mechanical resonance behavior of a ZnO nanowire/nanorod at ambient condition has been studied under optical microscope by cutting its length using focused ion beam microscopy. Nanobalance using a ZnO nanowire as the cantilever has been demonstrated for measuring the mass in the order of pico-grams in working atmosphere (see optical microscopy images). The measurement limit of the balance is estimated to be ∼1 pg. The technique demonstrated here has potential for commercial applications in general laboratories, especially for measuring the mass of wet biological cells or species.     

N型局部有源忆阻器的神经形态行为

局部有源忆阻器(locally-active memristor,LAM)凭借其高集成度、低功耗和局部有源特性等优点,在神经形态计算领域显示出巨大的潜力.本文提出了一种简单的N型LAM数学模型,通过揭示其非线性动力特性,设计了N型LAM神经元电路.采用Hopf分岔、数值分析等方法定量研究了该电路的动力学行为,成功模拟了多种神经形态行为,包括全或无行为、尖峰、簇发、周期振荡等.并利用该神经元电路结构模拟了生物触觉神经元的频率特性.仿真结果表明:当输入信号幅值低于阈值时,神经元电路输出信号的振荡频率与输入信号强度呈正相关(即兴奋状态),并在阈值处达到最大值.随后,继续增大激励强度,振荡频率则逐渐降低(即保护性抑制状态).最后,设计了N型LAM硬件仿真器,并完成了人工神经元电路的硬件实现,实验结果与仿真结果、理论分析相一致,验证了该N型LAM具备的神经形态行为.     

An electrical characterization of a hetero-junction nanowire (NW) PN diode (n-GaN NW/p-Si) formed by dielectrophoresis alignment

This is a report on the electrical characterization of gallium nitride (GaN) nanowire (NW) p–n junction diodes. These diodes were formed by assembling n-GaN NWs on p-Si (1 0 0) substrates using alternating current (AC) dielectrophoresis (DEP). The AC DEP was optimized with a bias voltage of 15 V_p−p at a frequency of 1 kHz. The hetero-junction single GaN nanowire p-n diode (n-GaN NW/p-Si) showed well-defined current rectifying behavior with a forward voltage drop of 1.2–2.0 V at a current density of 10–60 A/cm². The GaN nanowire p–n diodes had a high parasite resistance in the range of >470 kΩ. We observed that these high resistances were mostly the result of the metal contacts to the n-GaN NWs. We also found that these parasite resistances were reduced by the formation of an additional capping layer on the top of the n-GaN NW as well as high temperature annealing.     

梯度耦合下神经元网络中靶波和螺旋波的诱发研究

神经系统内数量众多的神经元电活动的群体行为呈现一定的节律性和自组织性. 当网络局部区域存在异质性或者受到持续周期性刺激, 则在网络内诱发靶波, 且这些靶波如'节拍器'可调制介质中行波的诱发和传播. 基于Hindmarsh-Rose 神经元模型构造了最近邻连接下的二维神经元网络, 研究在非均匀耦合下神经元网络内有序波的诱发问题. 在研究中, 选定网络中心区域的耦合强度最大, 从中心向边界的神经元之间的耦合强度则按照阶梯式下降. 研究结果表明, 在恰当的耦合梯度下, 神经元网络内诱发的靶波或螺旋波可以占据整个网络, 并有效调制神经元网络的群体电活动, 使得整个网络呈现有序性. 特别地, 当初始值为随机值时, 梯度耦合也可以诱发稳定的有序态. 这种梯度耦合对网络群体行为调制的研究结果有助于理解神经元网络的自组织行为.     

Scaling of degree correlations and its influence on diffusion in scale-free networks

Connectivity correlations play an important role in the structure of scale-free networks. While several empirical studies exist, there is no general theoretical analysis that can explain the largely varying behavior of real networks. Here, we use scaling theory to quantify the degree of correlations in the particular case of networks with a power-law degree distribution. These networks are classified in terms of their correlation properties, revealing additional information on their structure. For instance, the studied social networks and the Internet at the router level are clustered around the line of random networks, implying a strongly connected core of hubs. On the contrary, some biological networks and the WWW exhibit strong anticorrelations. The present approach can be used to study robustness or diffusion, where we find that anticorrelations tend to accelerate the diffusion process.     

Goos-Hänchen effect of spin electron beams in a parallel double <Emphasis Type="Italic">δ</Emphasis>-barrier magnetic nanostructure

In this paper, we study the effect of time-periodic coupling strength (TPCS) and network connection degree ⟨k⟩ on the temporal coherence of the chaotic bursting of the scale-free networks of thermo-sensitive neurons. It is found that the chaotic bursting becomes ordered and can exhibit coherence resonance (CR) when TPCS amplitude ε ₀ or the network connection degree ⟨k⟩ is varied. In particular, the neuronal bursting may exhibit multiple CR (MCR) behavior when TPCS frequency ω is varied. It is also found that, as ⟨k⟩ is increased, the value of ε ₀ for the MCR decreases, but the frequency for the MCR almost keeps unchanged. These results show that the chaotic bursting can be tamed and the bursting temporal coherence can be enhanced and even optimized by TPCS and network connection degree. Furthermore, TPCS can repetitively enhance and even optimize the temporal coherence of the neuronal bursting behavior. These findings may help to better understand the roles of TPCS and network connection degree for improving the time precision of the information processing in neuronal networks.     

Magnetic properties of amorphous β-FeOOH nanowire arrays

β-FeOOH nanowire arrays with diameters of 50–200 nm have been fabricated by electrochemical deposition using two-step anodic porous alumina templates. The as-prepared nanowires are homogeneous and have large aspect ratios. The selected area electron diffraction photo performed on a single wire was used to confirm the amorphous crystal structure further. The magnetic properties of these nanowire arrays were firstly investigated by using a SQUID magnetometry. The ZFC and FC studies show that these nanowire arrays exhibit spin-freezing phenomena at low temperature. The temperature-dependent magnetization curves show that the Neel transition temperatures are much lower than that of bulk material. Moreover, hysteresis was found at 5 K and the coercivities up to about 1500 Oe. The size-dependent magnetic properties were also investigated. These abnormal magnetic behaviours can be interpreted in terms of the amorphous crystal structure and the low dimensionality of the nanowire arrays.