Double-pattern associative memory neural network with pattern loop

A double-pattern associative memory neural network with “pattern loop” is proposed. It can store 2N bit bipolar binary patterns up to the order of 2^2N , retrieve part or all of the stored patterns which all have the minimum Hamming distance with input pattern, completely eliminate spurious patterns, and has higher storing efficiency and reliability than conventional associative memory. The length of a pattern stored in this associative memory can be easily extended from 2N to κN.     

Adaptive output feedback control for nonlinear time-delay systems using neural network

This paper extends the adaptive neural network （NN） control approaches to a class of unknown output feedback nonlinear time-delay systems. An adaptive output feedback NN tracking controller is designed by backstepping technique. NNs are used to approximate unknown functions dependent on time delay, Delay-dependent filters are introduced for state estimation. The domination method is used to deal with the smooth time-delay basis functions. The adaptive bounding technique is employed to estimate the upper bound of the NN approximation errors. Based on Lyapunov- Krasovskii functional, the semi-global uniform ultimate boundedness of all the signals in the closed-loop system is proved, The feasibility is investigated by two illustrative simulation examples.     

混沌在Hopfield联想记忆网络中的应用

将混沌应用到Hofield联想记忆网络中,利用混沌的遍历性和随机性等独特的性质,可以使待联想模式跳出伪模式的吸引域,而到达存储模式的吸引域内,从而解决了Hopfield网络在噪信比较高的情况下,联想成功率较低的问题。仿真结果证明了该方法的有效性。     

A dynamic associative memory system by adopting an amygdala model

Although several kinds of computational associative memory models and emotion models have been proposed since the last century, the interaction between memory and emotion is almost always neglected in these conventional models. This study constructs a dynamic memory system, named the amygdala-hippocampus model, which intends to realize dynamic auto-association and the mutual association of time-series patterns more naturally by adopting an emotional factor, i.e., the functional model of the amygdala given by Morén and Balkenius. The output of the amygdala is designed to control the recollection state of multiple chaotic neural networks (MCNN) in CA3 of the hippocampus-neocortex model proposed in our early work. The efficiency of the proposed association system is verified by computer simulation using several benchmark time-series patterns. This work was presented in part at the 13th International Symposium on Artificial Life and Robotics, Oita, Japan, January 31–February 2, 2008     

A property of neural networks of associative memory with replacing units

A memory capacity exists for artificial neural networks of associative memory. The addition of new memories beyond the capacity overloads the network system and makes all learned memories irretrievable (catastrophic forgetting) unless there is a provision for forgetting old memories. This article describes a property of associative memory networks in which a number of units are replaced when networks learn. In our network, every time the network learns a new item or pattern, a number of units are erased and the same number of units are added. It is shown that the memory capacity of the network depends on the number of replaced units, and that there exists a optimal number of replaced units in which the memory capacity is maximized. The optimal number of replaced units is small, and seems to be independent of the network size. This work was presented in part at the 12th International Symposium on Artificial Life and Robotics, Oita, Japan, January 25–27, 2007     

神经网络内模控制在非线性时滞系统中的应用研究

针对工业中常见的时滞现象,提出把内模控制方法和神经控制原理有效结合起来,利用一种改进RBF神经网络对被控对象的模型和控制器进行自适应辨识,通过对实验室电加热炉这种典型一阶滞后对象实验,仿真表明,所提出的方法具有良好的控制特性,在系统受到干扰或对象参数发生变化的情况下,仍然具有良好的自适应性和鲁棒稳定性。     

Pattern classification using a linear associative memory

Pattern classification is a very important image processing task. A typical pattern classification algorithm can be broken into two parts; first, the pattern features are extracted and, second, these features are compared with a stored set of reference features until a match is found. In the second part, usually one of the several clustering algorithms or similarity measures is applied. In this paper, a new application of linear associative memory (LAM) to pattern classification problems is introduced. Here, the clustering algorithms or similarity measures are replaced by a LAM matrix multiplication. With a LAM, the reference features need not be separately stored. Since the second part of most classification algorithms is similar, a LAM standardizes the many clustering algorithms and also allows for a standard digital hardware implementation. Computer simulations on regular textures using a feature extraction algorithm achieved a high percentage of successful classification. In addition, this classification is independent of topological transformations.     

Robustness of network flow control against disturbances and time-delay

This paper studies robustness of Kelly's source and link control laws in (J. Oper. Res. Soc. 49 (1998) 237) with respect to disturbances and time-delays. This problem is of practical importance because of unmodelled flows, and propagation and queueing delays, which are ubiquitous in networks. We first show L_p-stability, for p[1,∞], with respect to additive disturbances. We pursue L_∞-stability within the input-to-state stability (ISS) framework of Sontag (IEEE Trans. Automat. Control 34 (1989) 435), which makes explicit the vanishing effect of initial conditions. Next, using this ISS property and a loop transformation, we prove that global asymptotic stability is preserved for sufficiently small time-delays in forward and return channels. For larger delays, we achieve global asymptotic stability by scaling down the control gains as in Paganini et al. (Proceedings of 2001 Conference on Decision and Control, Orlando, FL, December 2001, pp. 185–190)     

相空间压缩法实现混沌神经网络联想记忆

给出了利用相空间压缩法控制混沌神经网络,使得网络能够收敛于存储的目标模式的充分条件和必要条件.通过数学分析,得到了相空间压缩控制方法中对应参数的上下限;并通过对仿真结果的分析,提出了通过改变相空间压缩控制方法中对应的参数来实现混沌神经网络联想记忆的新方法.以上结果均通过仿真得到验证.     

一类二阶时延神经网络的分岔分析和控制

讨论了一类二阶时延网络系统的非线性特性,应用线性化稳定性和分岔理论,提出了该系统从稳定到分岔的条件.结论指出利用延迟时间可以进行分岔控制、极限环幅值控制等,并给出了仿真的具体实例.     

多维联想记忆神经网络在图像回忆中的应用

多维联想记忆神经网络可以用来回忆灰度图像。投影算法是回忆算法中的一类。采用不规则凸多边形的笛卡儿积构成的凸集代替正多边形的笛卡儿积构成的凸集,前者比后者更紧凑。数值实例表明,应用前者回忆灰度图像要比应用后者回忆灰度图像得到的图像更清晰,回忆所花时间更短。     

Anti-synchronization of stochastic perturbed delayed chaotic neural networks

This paper studies the anti-synchronization of a class of stochastic perturbed chaotic delayed neural networks. By employing the Lyapunov functional method combined with the stochastic analysis as well as the feedback control technique, several sufficient conditions are established that guarantee the mean square exponential anti-synchronization of two identical delayed networks with stochastic disturbances. These sufficient conditions, which are expressed in terms of linear matrix inequalities (LMIs), can be solved efficiently by the LMI toolbox in Matlab. Two numerical examples are exploited to demonstrate the feasibility and applicability of the proposed synchronization approaches.     

基于动态结构自适应神经网络的不确定时滞系统跟踪控制

针对一类不确定时滞非线性系统,提出一种自适应跟踪控制器.首先采用Lyapunov-Krasovskii函数设计时滞补偿器,并构造其中的参数调节规律.再针对建模误筹及小确定非线性,引入动态结构自适应神经网络,其隐层神经元个数可以随着跟踪误差的增大而在线增加,以提高逼近精度.最后,用仿真示例表明本文所提方法是有效的.     

Adaptive neural network dynamic surface control for perturbed nonlinear time-delay systems

This paper proposes an adaptive neural network control method for a class of perturbed strict-feedback nonlinear systems with unknown time delays. Radial basis function neural networks are used to approximate unknown intermediate control signals. By constructing appropriate Lyapunov-Krasovskii functionals, the unknown time delay terms have been compensated. Dynamic surface control technique is used to overcome the problem of "explosion of complexity" in backstepping design procedure. In addition, the semiglobal uniform ultimate boundedness of all the signals in the closed-loop system is proved. A main advantage of the proposed controller is that both problems of "curse of dimensionality" and "explosion of complexity" are avoided simultaneously. Finally, simulation results are presented to demonstrate the effectiveness of the approach.     

基于离散Hopfield神经网络的数字识别实现

介绍了离散Hopfield神经网络的基本概念;以MATLAB为工具,根据Hopfield神经网络的相关知识,设计了一个具有联想记忆功能的离散型Hopfield神经网络,并给出了设计思路、设计步骤和测试结果。实验结果表明,通过联想记忆,对于带有一定噪声的数字点阵,Hopfield网络可以正确地进行识别,且当噪声强度为0．1时的识别效果较好。     

非线性时滞系统的稳定自适应控制

针对模型不确定性的连续时间时滞系统，提出了一种新的神经网络自适应控制。系统的辨识模型是由神经网络和系统的已知信息组合构成，在此基础上，建立时滞系统的预测模型。基于神经网络预测模型的自适应控制器能够实现期望轨线的跟踪，理论上证明了闭环系统的稳定性。连续搅拌釜式反应器仿真结果表明了该控制方案的有效性。     

Analysis and design of associative memories based on recurrent neural network with discontinuous activation functions

This paper considers a recurrent neural network (RNN) with a special class of discontinuous activation function which is piecewise constants in the state space. One sufficient condition is established to ensure that the novel recurrent neural networks can have (4k−1)ⁿ locally exponential stable equilibrium points. Such RNN is suitable for synthesizing high-capacity associative memories. The design procedure is presented with the method of singular value decomposition. Finally, the validity and performance of the results are illustrated by use of two numerical examples.     

Chaos and hyperchaos in a Hopfield neural network

In this paper we investigate numerically the parameter-space of an autonomous system of four nonlinear first-order ordinary differential equations, which represents a Hopfield neural network with four neurons. The study considers three independent two-dimensional cross-sections of the three-dimensional parameter-space generated by this mathematical model, every constructed considering Lyapunov exponent values. We show that is possible to completely characterize the dynamics of the system based in these three plots, which are representative of the three-dimensional parameter-space as a whole.     

Identification and control of nonlinear systems by a time-delay recurrent neural network

In this paper, we first present a novel time-delay recurrent neural network (TDRNN) model by introducing the time-delay and recurrent mechanism. The proposed TDRNN model has special advantages such as simple structure, deeper depth and higher resolution ratio in memory. Thereafter, we develop the dynamic recurrent back-propagation algorithm for the TDRNN. To guarantee the fast convergence, the optimal adaptive learning rates are also derived in the sense of discrete-type Lyapunov stability. More specifically, a TDRNN identifier and a TDRNN controller are constructed to perform the identification and control of the nonlinear systems. Numerical experiments show that the TDRNN model has good effectiveness in the identification and control for dynamic systems.     

Global stability of bidirectional associative memory neural networks with continuously distributed delays

Global asymptotic stability of the equilibrium point of bidirectional associative memory (BAM) neural networks with continuously distributed delays is studied. Under two mild assumptions on the activation functions, two sufficient conditions ensuring global stability of such networks are derived by utilizing Lyapunov functional and some inequality analysis technique. The results here extend some previous results. A numerical example is given showing the validity of our method.