Optimal neural network algorithm for on-line string matching

We consider an online string matching problem in which we find all the occurrences of a pattern of m characters in a text of n characters, where all the characters of the pattern are available before processing, while the characters of the text are input one after the other. We propose a space-time optimal parallel algorithm for this problem using a neural network approach, This algorithm uses m McCulloch-Pitts neurons connected as a linear array. It processes every input character of the text in one step and hence it requires at most n iteration steps.     

Experimental study on arranging music by chaotic neural network

Based on the recalling ability on dynamic (chaotic) associative memory of neural networks, we have proposed two methods for making variations of an original melody. By computer simulations, we have shown candidates for the variation of the original melody taken from the first 16 bars of Minuet G major by Bach. The results obtained in this article may suggest a possibility that chaotic neural networks can excuse such a creative task as making variations of an original melody. © 1997 John Wiley & Sons, Inc.     

Generating novel memories by integration of chaotic neural network modules

A principle of integrating neural network modules based on chaotic dynamics was studied on our two-moduled Nozawa model. Chaotic neural networks represent each embedded pattern as a low-dimensional periodic orbit, and the others are shown as high-dimensional chaotic attractors. This is equivalent to W. Freeman’s “I don’t know” and “I know” states. In particular, we noted that the combination of two-way inputs to each neural network module conflicted with embedded Hebbian correspondence. It was found that the interaction between the modules generated a novel “I know” state in addition to the embedded representation. Chaotic neural network modules can autonomously generate novel memories or functions by this interaction. The result suggests a functional integration in neural networks as it ought to be, e.g., feature binding and gestalt. This work was presented, in part, at the Fourth International Symposium on Artificial Life and Robotics, Oita, Japan, January 19–22, 1999     

一种用于函数优化的小波混沌神经网络

混沌神经网络能有效地解决函数优化问题。通过把sigmoid函数转化为墨西哥帽小波函数,而单一化退火因子函数被分段指数模拟退火函数所取代,提出了一种新型的混沌神经网络。与传统的混沌神经网络相比,该网络具有更强的全局寻优能力。仿真结果表明,小波混沌神经网络在搜索全局最优解的速度和精确度上都明显优于传统的混沌神经网络。     

Local stereovision matching through the ADALINE neural network

This paper presents an approach to the local stereovision matching problem using edge segments as features with four attributes. Based on these attributes we compute a matching probability between pairs of features of the stereo images. A correspondence is said to be true when this probability is maximum. The probability value is a weighted sum of the attributes. We use two combined ADALINE neural networks to compute the weight for each attribute. A comparative analysis among other recent matching methods is illustrated.     

Hysteresis modeling based on the hysteretic chaotic neural network

The hysteresis activation function is proposed, and a novel hysteretic chaotic neuron model is constructed by the function. It is shown that the model may exhibit a complex dynamic behavior. On the basis of this neuron model, we propose a novel neural network, which can be applied to hysteresis system modeling. We demonstrate the advantages of the network by experimental results.     

On-demand chaotic neural network for broadcast scheduling problem

This paper presents a novel approach to optimizing network packet transfer scheme through introducing a new method for on-demand chaotic noise injection strategy for the Broadcast Scheduling Problem (BSP). Packet radio networks have many applications, while finding an optimized scheduling to transmit data is proven to be a NP-hard problem. The objective of the proposed method is to find an optimal time division multiple access (TDMA) frame, based on maximizing the channel utilization. The proposed method benefits from an on-demand noise injection policy, which injects noise based on the status of neuron and its neighborhoods. The method is superior to other Noise Chaotic Neural Networks (NCNN) that suffer from blind injection policy. The experimental result shows that, in most cases, the proposed on-demand noise injection algorithm finds the best solution with minimal average time delay and maximum channel utilization.     

A neural network system for matching dental radiographs

This paper addresses the problem of creating a postmortem identification system by matching image features extracted from dental radiographs. We lay the architecture of a prototype automated dental identification system (ADIS), which tackles the dental image matching problem by first extracting high-level features to expedite retrieval of potential matches and then by low-level image comparison using inherent features of dental images. We propose the use of learnable inherent dental image features for tooth-to-tooth image comparisons. We treat the tooth-to-tooth matching problem as a binary classification problem for which we propose probabilistic models of class-conditional densities. We also propose an adaptive strategic searching technique and use it in conjunction with back propagation in order to estimate system parameters. We present promising experimental results that reflect the value of our approach.     

基于DSP的混沌BP神经网络的实现

文中介绍了一种将混沌和BP神经网络结合起来的算法,以解决由于采用梯度算法使得标准BP神经网络存在易陷入局部极小值的问题。并提出一种利用TI公司的TMS320C5410芯片实现混沌BP神经网络算法的软硬件方案。该方案具有运算能力强、实时性好的特点。     

Graph matching by neural relaxation

We propose a new relaxation scheme for graph matching in computer vision. The main distinguishing feature of our approach is that matching is formulated as a process of eliminating unlikely candidates rather than finding the best match directly. Bayesian development leads to a robust algorithm which can be implemented in a fast and efficient manner on a neural network architecture. We illustrate the utility of the technique through comparisons with its conventional counterpart on simulated and real-world data.     

A novel Hash algorithm construction based on chaotic neural network

An algorithm for constructing a one-way novel Hash function based on two-layer chaotic neural network structure is proposed. The piecewise linear chaotic map (PWLCM) is utilized as transfer function, and the 4-dimensional and one-way coupled map lattices (4D OWCML) is employed as key generator of the chaotic neural network. Theoretical analysis and computer simulation indicate that the proposed algorithm presents several interesting features, such as high message and key sensitivity, good statistical properties, collision resistance and secure against meet-in-the-middle attacks, which can satisfy the performance requirements of Hash function.     

An artificial neural network noise reduction method for chaotic attractors

A method is presented to reduce noise in chaotic attractors without knowing the underlying maps. The method is based on using Artificial Neural Network (ANN) for moderate levels of additive noise. For high levels of additive noise, a combination of a refinement procedure with ANN is used. In this case, only one refinement is needed for the successful use of ANN. The obtained ANN model is used for long-term predictions of the future behavior of a Henon attractor, using information based only on past values.     

Parallel chaotic Hash function construction based on cellular neural network

A new parallel chaotic Hash function, based on four-dimensional cellular neural network, is proposed in this paper. The message is expanded by iterating chaotic logistic map and then divided into blocks with a length of 512 bits each. All blocks are processed in a parallel mode, which is one of the significant characteristics of the proposed algorithm. Each 512-bit block is divided into four 128-bit sub-blocks, each of which is further separated into four 32-bit values and then the four values are mixed into four new values generated by chaotic cat map. The obtained four new values are performed by the bit-wise exclusive OR operation with four initial values or previously generated four values, and then, they are used as the inputs of cellular neural network. By iterating cellular neural network, another four values as the middle Hash value are generated. The generated values of all blocks are inputted into the compression function to produce the final 128-bit Hash value. Theoretical analysis and computer simulation indicate that the proposed algorithm satisfies the requirements of a secure Hash function.     

A noisy chaotic neural network for solving combinatorial optimization problems: stochastic chaotic simulated annealing

Recently Chen and Aihara have demonstrated both experimentally and mathematically that their chaotic simulated annealing (CSA) has better search ability for solving combinatorial optimization problems compared to both the Hopfield-Tank approach and stochastic simulated annealing (SSA). However, CSA may not find a globally optimal solution no matter how slowly annealing is carried out, because the chaotic dynamics are completely deterministic. In contrast, SSA tends to settle down to a global optimum if the temperature is reduced sufficiently slowly. Here we combine the best features of both SSA and CSA, thereby proposing a new approach for solving optimization problems, i.e., stochastic chaotic simulated annealing, by using a noisy chaotic neural network. We show the effectiveness of this new approach with two difficult combinatorial optimization problems, i.e., a traveling salesman problem and a channel assignment problem for cellular mobile communications.     

The agile improvement of MMORPGs based on the enhanced chaotic neural network

Massive multiplayer online role playing games (MMORPGs) should be continuously and incrementally modified to meet with the customers’ increasing demand. However, that is a tough task since the number of customers is huge and all of the customers are geographically dispersed. Therefore, this paper presents an agile improvement system by using the action diagram and the enhanced chaotic neural network (ECNN) model, combining the merits of chaotic neural network, fuzzy analytic hierarchy process (AHP), and genetic algorithm in one consolidated model. It is expected that the proposed method will overcome most of the disadvantages of published models, particularly the accuracy of customer satisfaction model and the validity of modification decision. Also, it gives a chance to meet the demands of customers at an optimal cost and make the hard but necessary improvement decisions whenever they are required.     

Learning chaotic attractors by neural networks

An algorithm is introduced that trains a neural network to identify chaotic dynamics from a single measured time series. During training, the algorithm learns to short-term predict the time series. At the same time a criterion, developed by Diks, van Zwet, Takens, and de Goede (1996) is monitored that tests the hypothesis that the reconstructed attractors of model-generated and measured data are the same. Training is stopped when the prediction error is low and the model passes this test. Two other features of the algorithm are (1) the way the state of the system, consisting of delays from the time series, has its dimension reduced by weighted principal component analysis data reduction, and (2) the user-adjustable prediction horizon obtained by "error propagation"-partially propagating prediction errors to the next time step. The algorithm is first applied to data from an experimental-driven chaotic pendulum, of which two of the three state variables are known. This is a comprehensive example that shows how well the Diks test can distinguish between slightly different attractors. Second, the algorithm is applied to the same problem, but now one of the two known state variables is ignored. Finally, we present a model for the laser data from the Santa Fe time-series competition (set A). It is the first model for these data that is not only useful for short-term predictions but also generates time series with similar chaotic characteristics as the measured data.     

用综合法优化前向神经网络结构

在神经网络研究中,如何确定神经网的结构是一个重要的研究方向.提出了一种通用的确定前向神经网络结构的自适应方法,即先用动态增长法快速训练网络拓扑结构及权值至满足给定的误差为止,然后用遗传算法(GA)对训练好的网络剪枝.实验表明,算法具有较好的通用性和可扩展性,收敛速度较快,对进一步的数据挖掘具有重要的意义.     

Dynamic characteristic of a multiple chaotic neural network and its application

Based on chaotic neural network, a multiple chaotic neural network algorithm combining two different chaotic dynamics sources in each neuron is proposed. With the effect of self-feedback connection and non-linear delay connection weight, the new algorithm can contain more powerful chaotic dynamics to search the solution domain globally in the beginning searching period. By analyzing the dynamic characteristic and the influence of cooling schedule in simulated annealing, a flexible parameter tuning strategy being able to promote chaotic dynamics convergence quickly is introduced into our algorithm. We show the effectiveness of the new algorithm in two difficult combinatorial optimization problems, i.e., a traveling salesman problem and a maximum clique problem.     

基于混沌神经网络的故障暂态量选线方法研究

为了有效提取故障暂态信息,研究选取适当的小波包基函数。针对电力系统故障暂态量的特点,为了有效克服非故障暂态信号的干扰,研究选取了容错性和联想记忆功能很强的混沌神经网络实现故障选线,并采用改进的遗传算法对混沌神经网络的权值和参数同时进行训练,加快其收敛速度。根据目标模式与神经元的输出状态构造了数值型选线判据。通过实验算例验证了基于暂态量选线判据的有效性和适用性。     

A transiently chaotic neural-network implementation of the CDMA multiuser detector

The complex dynamics of the chaotic neural networks makes it possible for them to escape from local minimum of the simple gradient descent neurodynamics. We use a transiently chaotic neural network to detect the CDMA multiuser signals and hence obtain an implementation scheme of the CDMA multiuser detector (TCNN-MD). Computer simulation results show that the proposed detector is clearly superior to Hopfield neural-network-based detector.