Event-driven simulation of spiking neurons with stochastic dynamics

We present a new technique, based on a proposed event-based strategy (Mattia & Del Giudice, 2000), for efficiently simulating large networks of simple model neurons. The strategy was based on the fact that interactions among neurons occur by means of events that are well localized in time (the action potentials) and relatively rare. In the interval between two of these events, the state variables associated with a model neuron or a synapse evolved deterministically and in a predictable way. Here, we extend the event-driven simulation strategy to the case in which the dynamics of the state variables in the inter-event intervals are stochastic. This extension captures both the situation in which the simulated neurons are inherently noisy and the case in which they are embedded in a very large network and receive a huge number of random synaptic inputs. We show how to effectively include the impact of large background populations into neuronal dynamics by means of the numerical evaluation of the statistical properties of single-model neurons under random current injection. The new simulation strategy allows the study of networks of interacting neurons with an arbitrary number of external afferents and inherent stochastic dynamics.     

Spontaneous dynamics of asymmetric random recurrent spiking neural networks

In this letter, we study the effect of a unique initial stimulation on random recurrent networks of leaky integrate-and-fire neurons. Indeed, given a stochastic connectivity, this so-called spontaneous mode exhibits various nontrivial dynamics. This study is based on a mathematical formalism that allows us to examine the variability of the afterward dynamics according to the parameters of the weight distribution. Under the independence hypothesis (e.g., in the case of very large networks), we are able to compute the average number of neurons that fire at a given time-the spiking activity. In accordance with numerical simulations, we prove that this spiking activity reaches a steady state. We characterize this steady state and explore the transients.     

Improved Izhikevich neurons for spiking neural networks

Spiking neural networks constitute a modern neural network paradigm that overlaps machine learning and computational neurosciences. Spiking neural networks use neuron models that possess a great degree of biological realism. The most realistic model of the neuron is the one created by Alan Lloyd Hodgkin and Andrew Huxley. However, the Hodgkin–Huxley model, while accurate, is computationally very inefficient. Eugene Izhikevich created a simplified neuron model based on the Hodgkin–Huxley equations. This model has better computational efficiency than the original proposed by Hodgkin and Huxley, and yet it can successfully reproduce all known firing patterns. However, there are not many articles dealing with implementations of this model for a functional neural network. This study presents a spiking neural network architecture that utilizes improved Izhikevich neurons with the purpose of evaluating its speed and efficiency. Since the field of spiking neural networks has reinvigorated the interest in biological plausibility, biological realism was an additional goal. The network is tested on the correct classification of logic gates (including XOR) and on the iris dataset. Results and possible improvements are also discussed.     

Vectorized algorithms for spiking neural network simulation

High-level languages (Matlab, Python) are popular in neuroscience because they are flexible and accelerate development. However, for simulating spiking neural networks, the cost of interpretation is a bottleneck. We describe a set of algorithms to simulate large spiking neural networks efficiently with high-level languages using vector-based operations. These algorithms constitute the core of Brian, a spiking neural network simulator written in the Python language. Vectorized simulation makes it possible to combine the flexibility of high-level languages with the computational efficiency usually associated with compiled languages.     

System dynamics simulation of Hopfield neural networks

Hopfield networks are a class of neural network models where non-linear graded response neurons organized into networks with effectively symmetric synaptic connections are able to implement interesting algorithms, thereby introducing the concept of information storage in the stable states of dynamical systems. In addition to opening up the possibility of using system dynamics as a vehicle to gain potentially useful insights into the behaviour of such networks, especially in the field or nonelectrical engineering, we study the dynamics of the state-space trajectory as well as time domain evolution of sensitivities of the states with respect to circuit parameters.     

面向类脑计算的脉冲神经网络研究

随着深度学习在训练成本、泛化能力、可解释性以及可靠性等方面的不足日益突出,类脑计算已成为下一代人工智能的研究热点。脉冲神经网络能更好地模拟生物神经元的信息传递方式,且具有计算能力强、功耗低等特点,在模拟人脑学习、记忆、推理、判断和决策等复杂信息方面具有重要的潜力。本文对脉冲神经网络从以下几个方面进行总结:首先阐述脉冲神经网络的基本结构和工作原理;在结构优化方面,从脉冲神经网络的编码方式、脉冲神经元改进、拓扑结构、训练算法以及结合其他算法这5个方面进行总结;在训练算法方面,从基于反向传播方法、基于脉冲时序依赖可塑性规则方法、人工神经网络转脉冲神经网络和其他学习算法这4个方面进行总结;针对脉冲神经网络的不足与发展,从监督学习和无监督学习两方面剖析;最后,将脉冲神经网络应用到类脑计算和仿生任务中。本文对脉冲神经网络的基本原理、编码方式、网络结构和训练算法进行了系统归纳,对脉冲神经网络的研究发展具有一定的积极意义。     

Modeling honeybee communication using network of spiking neural networks to simulate nectar reporting behavior

The paper presents the findings of the research that attempted to mathematically model the cognitive behavior that could arise due to the interaction between honeybees in a colony during forager recruitment process. The model defines a honeybee as a spiking neural network, and colony as a network of spiking neural networks. The proposed mathematical model has been evaluated by analyzing the cognitive behavior generated by the main network which represents honeybees’ interaction as interactions of component networks (i.e. spiking neural networks). Accordingly, behavior of the component network, that represents an unemployed forager in the colony, was examined under different scenarios by setting networks’ parameters to simulate ecological situations in the colony. The reporting of different level of quantity of nectar sources by scouts to the colony, an attempt made by a scout to attract more unemployed foragers for foraging, and influence by dancing foragers to attract other unemployed foragers for foraging are those ecological colony states that have been tested in this research. The results of all these cases have supported that the proposed mathematical model can sufficiently simulate the unemployed forager’s behavior during recruitment process.     

Layered tile architecture for efficient hardware spiking neural networks

Spiking Neural Network (SNN) is the most recent computational model that can emulate the behaviour of biological neuron system. However, its main drawback is that it is computationally intensive, which limits the system scalability. This paper highlights and discusses the importance and significance of emulating SNNs in hardware devices. A layer-level tile architecture (LTA) is proposed for hardware-based SNNs. The LTA employs a two-level sharing mechanism of computing components at the synapse and neuron levels, and achieves a trade-off between computational complexity and hardware resource costs. The LTA is implemented on a Xilinx FPGA device. Experimental results demonstrate that this approach is capable of scaling to large hardware-based SNNs.     

Efficient and hardware-friendly methods to implement competitive learning for spiking neural networks

Neural Computing and Applications - Spiking neural network (SNN) trained by spike-timing-dependent plasticity (STDP) is a promising computing paradigm for energy-efficient artificial intelligence...     

A remote password authentication scheme for multiserverarchitecture using neural networks

Conventional remote password authentication schemes allow a serviceable server to authenticate the legitimacy of a remote login user. However, these schemes are not used for multiserver architecture environments. We present a remote password authentication scheme for multiserver environments. The password authentication system is a pattern classification system based on an artificial neural network. In this scheme, the users only remember user identity and password numbers to log in to various servers. Users can freely choose their password. Furthermore, the system is not required to maintain a verification table and can withstand the replay attack.     

Detection of weather images by using spiking neural networks of deep learning models

Neural Computing and Applications - The transmission of weather information of a location at certain time intervals affects the living conditions of the people there directly or indirectly....     

Dynamic simulation of vapor-compression cycle using neural networks

In this paper, the thermodynamic modeling of a vapor-compression cycle, based on a neural network approach is presented. A generalized radial basis function is used for the network, which takes previous control inputs and previous states as the network input and generates the predicted current state as the network output. The trained network is validated by non-trained data and shows all the process characteristics of a vapor-compression cycle for an air-to-water heat pump to a satisfactory degree.     

基于卷积计算的多层脉冲神经网络的监督学习

针对脉冲神经元基于精确定时的多脉冲编码信息的特点,提出了一种基于卷积计算的多层脉冲神经网络监督学习的新算法。该算法应用核函数的卷积计算将离散的脉冲序列转换为连续函数,在多层前馈脉冲神经网络结构中,使用梯度下降的方法得到基于核函数卷积表示的学习规则,并用来调整神经元连接的突触权值。在实验部分,首先验证了该算法学习脉冲序列的效果,然后应用该算法对Iris数据集进行分类。结果显示,该算法能够实现脉冲序列复杂时空模式的学习,对非线性模式分类问题具有较高的分类正确率。     

Adaptive consensus protocol for networks of multiple agents with nonlinear dynamics using neural networks

In this paper, an adaptive protocol is proposed to solve the consensus problem of multi‐agent systems with high‐order nonlinear dynamics by using neural networks (NNs) to approximate the unknown nonlinear system functions. It is derived that all agents achieve consensus if the undirected interaction graph is connected, and the transient performance of the multi‐agent system is also investigated. It shows that the adaptive protocol and the consensus analysis can be easily extended to switching networks by the existing LaSalle's Invariance Principle of switched systems. A numerical simulation illustrates the effectiveness of the proposed consensus protocol. Copyright © 2011 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society     

Correction to: Robust facial landmark extraction scheme using multiple convolutional neural networks

Multimedia Tools and Applications - In the original publication, the author name “Seungmin Rho” was incorrectly spelled as “Seumgmin Rho”. The correct author name is given...     

Heuristic pattern correction scheme using adaptively trainedgeneralized regression neural networks

In many pattern classification problems, an intelligent neural system is required which can learn the newly encountered but misclassified patterns incrementally, while keeping a good classification performance over the past patterns stored in the network. In the paper, an heuristic pattern correction scheme is proposed using adaptively trained generalized regression neural networks (GRNNs). The scheme is based upon both network growing and dual-stage shrinking mechanisms. In the network growing phase, a subset of the misclassified patterns in each incoming data set is iteratively added into the network until all the patterns in the incoming data set are classified correctly. Then, the redundancy in the growing phase is removed in the dual-stage network shrinking. Both long- and short-term memory models are considered in the network shrinking, which are motivated from biological study of the brain. The learning capability of the proposed scheme is investigated through extensive simulation studies.     

Robust facial landmark extraction scheme using multiple convolutional neural networks

Multimedia Tools and Applications - Facial landmarks are a set of features that can be distinguished on the human face with the naked eye. Typical facial landmarks include eyes, eyebrows, nose, and...     

Copyright-proving scheme for audio with counter-propagation neural networks

Protecting the intellectual property rights (IPR) of digital media has become an important issue. In this paper, counter-propagation neural networks (CPN) are applied to audio copyright protection. The db4 filter of the Daubechies wavelet is used on the original audio signals. The coefficients obtained from the 4-level Daubechies (db4) filter and the corresponding copyright information are used to train the CPN. Since the low-frequency components of DWT are robust, most noise is excluded. The CPN has memory and fault tolerance, so most attacks do not degrade the quality of the extracted copyright image. Moreover, the copyright generation procedure and the verification procedure are integrated into the proposed CPN method. Experimental results show that the proposed CPN-based method is robust, inaudible, and authentic.     

Hybrid accident simulation methodology using artificial neural networks for nuclear power plants

A hybrid accident simulation methodology for nuclear power plants is proposed to enhance the capabilities of compact simulator by introducing artificial neural networks. Two neural networks are trained with the target values obtained from the analyses of detailed computer codes and trained results are combined with the compact simulator to perform the following roles: (i) compensation for inaccuracies of a compact simulator occurring from simplified governing equation and reduced number of physical control volumes, and (ii) prediction of the critical parameter usually calculated from the sophisticated computer code: the autoassociative neural network improves the computational results of the compact simulator up to the accuracy level of detailed best estimate computer code, while the backpropagation neural network predicts the minimum departure from nucleate boiling ratio (DNBR). Simulations are carried out to verify the applicability of the proposed methodology for the loss of flow accidents and the results show that the neural networks can be used as a complementary tool to improve the results of a compact simulator.     

Artificial neural networks for job shop simulation

This paper explores the use of artificial neural networks (ANNs) as a valid alternative to the traditional job-shop simulation approach. Feed forward, multi-layered neural network metamodels were trained through the back-error-propagation (BEP) learning algorithm to provide a versatile job-shop scheduling analysis framework. The constructed neural network architectures were capable of satisfactorily estimating the manufacturing lead times (MLT) for orders simultaneously processed in a four-machine job shop. The MLTs produced by the developed ANN models turned out to be as valid as the data generated from three well-known simulation packages, i.e. Arena, SIMAN, and ProModel. The ANN outputs proved not to be substantially different from the results provided by other valid models such as SIMAN and ProModel when compared against the adopted baseline, Arena. The ANN-based simulations were able to fairly capture the underlying relationship between jobs' machine sequences and their resulting average flowtimes, which proves that ANNs are a viable tool for stochastic simulation metamodeling.