A neural network approach for approximate force response analyses of a bridge population

In this paper, artificial neural networks (ANNs) are used to develop an efficient method for rapid and approximate force response analyses of a bridge population. The single-span reinforced concrete T-beam bridge population in Pennsylvania State is taken as a particular case study. First, a statistical analysis is conducted to examine implicit and explicit dependencies between various geometrical and structural parameters of the bridges, and the governing bridge parameters are identified along with their ranges of variation within the population. Then, a set of sample bridges are randomly generated using different combinations of the governing parameters within their predefined ranges of variation. An exact finite element analysis is implemented for each sample bridge, and the maximum moment and shear responses in beams are obtained at critical locations under various combinations of standard truck loads. An ANN is implemented to learn the relationship between the bridge parameters (inputs) and responses (outputs) based on the sample set and to make predictions for other bridges that are not present in the set. The performances of a variety of different ANN architectures are tested, and their prediction capabilities are measured and compared.     

Effective data mining using neural networks

Classification is one of the data mining problems receiving great attention recently in the database community. The paper presents an approach to discover symbolic classification rules using neural networks. Neural networks have not been thought suited for data mining because how the classifications were made is not explicitly stated as symbolic rules that are suitable for verification or interpretation by humans. With the proposed approach, concise symbolic rules with high accuracy can be extracted from a neural network. The network is first trained to achieve the required accuracy rate. Redundant connections of the network are then removed by a network pruning algorithm. The activation values of the hidden units in the network are analyzed, and classification rules are generated using the result of this analysis. The effectiveness of the proposed approach is clearly demonstrated by the experimental results on a set of standard data mining test problems     

MerIt: improving neural program synthesis by merging collective intelligence

Automated Software Engineering - Identifying security concerns is a security activity that can be integrated into the requirements development phase. However, it has been shown that manually...     

An orthogonal neural network for function approximation

This paper presents a new single-layer neural network which is based on orthogonal functions. This neural network is developed to avoid the problems of traditional feedforward neural networks such as the determination of initial weights and the numbers of layers and processing elements. The desired output accuracy determines the required number of processing elements. Because weights are unique, the training of the neural network converges rapidly. An experiment in approximating typical continuous and discrete functions is given. The results show that the neural network has excellent performance in convergence time and approximation error.     

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。     

A note on quantum correlations in Werner states under two collective noises

Influences of two collective noises, i.e., the dephasing noise and the rotating noise, on quantum correlations in Werner states are considered in this note. It is found that the collective noises do not alter the correlations as far as quantum discord and geometry discord as quantifiers are concerned. Alternatively, quantum correlations in Werner states are robust against the noises.     

Fault-tolerant controlled deterministic secure quantum communication using EPR states against collective noise

This paper proposes two new fault-tolerant controlled deterministic secure quantum communication (CDSQC) protocols based only on Einstein–Podolsky–Rosen (EPR) entangled states. The proposed protocols are designed to be robust against the collective-dephasing noise and the collective-rotation noise, respectively. Compared to the existing fault-tolerant controlled quantum communication protocols, the proposed protocols not only can do without a quantum channel between the receiver and the controller as the state-of-the-art protocols do, but also have the advantage that the number of quantum particles required in the CDSQC protocols is reduced owing to the use of the simplest entangled states.     

An energy function for the random neural network

Since Hopfield's seminal work on energy functions for neural networks and their consequence for the approximate solution of optimization problems, much attention has been devoted to neural heuristics for combinatorial optimization. These heuristics are often very time-consuming because of the need for randomization or Monte Carlo simulation during the search for solutions. In this paper, we propose a general energy function for a new neural model, the random neural model of Gelenbe. This model proposes a scheme of interaction between the neurons and not a dynamic equation of the system. Then, we apply this general energy function to different optimization problems.     

Taking advantage of collective knowledge in emergency response systems

When large groups work on a theme, they have the potential to produce a lot of useful knowledge, regardless of whether they are acting in a coordinated manner or individually. Spontaneously generated information has received much attention in recent years, as organizations and businesses discover the power of crowds. New technologies, such as blogs, Twitter, wikis, photo sharing, collaborative tagging and social networking sites, enable the creation and dissemination of content in a relatively simple way. As a result, the aggregate body of knowledge is growing at an accelerated rate. Many organizations are looking for ways to harness this power, which is being called collective intelligence. Research has shown that it is possible to obtain high quality results from collectively produced work.In this paper, we consider the domain of emergency response. Research has shown that individuals respond quickly and massively to emergencies, and that they try to help with the situation. Thus, it seems like a logical step to attempt to harness collective knowledge for emergency management. Disaster relief groups and field command frequently suffer from lack of up to date information, which may be critical in a rapidly evolving situation. Some of this information could be generated by the crowd at large, enabling more effective response to the situation. In this paper, we discuss the possibilities for the introduction of collective knowledge in disaster relief and present architecture and examples of how this could be accomplished.     

On global-local artificial neural networks for function approximation

We present a hybrid radial basis function (RBF) sigmoid neural network with a three-step training algorithm that utilizes both global search and gradient descent training. The algorithm used is intended to identify global features of an input-output relationship before adding local detail to the approximating function. It aims to achieve efficient function approximation through the separate identification of aspects of a relationship that are expressed universally from those that vary only within particular regions of the input space. We test the effectiveness of our method using five regression tasks; four use synthetic datasets while the last problem uses real-world data on the wave overtopping of seawalls. It is shown that the hybrid architecture is often superior to architectures containing neurons of a single type in several ways: lower mean square errors are often achievable using fewer hidden neurons and with less need for regularization. Our global-local artificial neural network (GL-ANN) is also seen to compare favorably with both perceptron radial basis net and regression tree derived RBFs. A number of issues concerning the training of GL-ANNs are discussed: the use of regularization, the inclusion of a gradient descent optimization step, the choice of RBF spreads, model selection, and the development of appropriate stopping criteria.     

Sigma function in observer design for states and perturbations

For nonlinear systems operating under uncertainty, this paper involves the principle of motion separation to design a state observer with nonlinear corrections in the form of sigma functions. For the systems representable in the regular form with respect to the external perturbations, the above approach yields the current estimates of the unmeasurable state variables and external perturbations without extending the dynamic order of the observer by a model that simulates the action of the external perturbations. The developed algorithms are applied in the control system of an asynchronous drive with an incomplete set of measuring devices.     

A low-complexity fuzzy activation function for artificial neural networks

A novel fuzzy-based activation function for artificial neural networks is proposed. This approach provides easy hardware implementation and straightforward interpretability in the basis of IF-THEN rules. Backpropagation learning with the new activation function also has low computational complexity. Several application examples ( XOR gate, chaotic time-series prediction, channel equalization, and independent component analysis) support the potential of the proposed scheme.     

Wavelet basis function neural networks for sequential learning.

In this letter, we develop the wavelet basis function neural networks (WBFNNs). It is analogous to radial basis function neural networks (RBFNNs) and to wavelet neural networks (WNNs). In WBFNNs, both the scaling function and the wavelet function of a multiresolution approximation (MRA) are adopted as the basis for approximating functions. A sequential learning algorithm for WBFNNs is presented and compared to the sequential learning algorithm of RBFNNs. Experimental results show that WBFNNs have better generalization property and require shorter training time than RBFNNs.     

Constructive neural networks with piecewise interpolationcapabilities for function approximations

This paper proposes a constructive neural network with a piecewise linear or nonlinear local interpolation capability to approximate arbitrary continuous functions. This neural network is devised by introducing a space tessellation which is a covering of the Euclidean space by nonoverlapping hyperpolyhedral convex cells. In the proposed neural network, a number of neural network granules (NNG's) are processed in parallel and repeated regularly with the same structures. Each NNG does a local mapping with an interpolation capability for a corresponding hyperpolyhedral convex cell in a tessellation. The plastic weights of the NNG can be calculated to implement the mapping for training data; consequently, this reduces training time and alleviates the difficulties of local minima in training. In addition, the interpolation capability of the NNG improves the generalization for the new data within the convex cell. The proposed network requires additional neurons for tessellation over the standard multilayer neural networks. This increases the network size but does not slow the retrieval response when implemented by parallel architecture.     

A novel activation function for multilayer feed-forward neural networks

Traditional activation functions such as hyperbolic tangent and logistic sigmoid have seen frequent use historically in artificial neural networks. However, nowadays, in practice, they have fallen out of favor, undoubtedly due to the gap in performance observed in recognition and classification tasks when compared to their well-known counterparts such as rectified linear or maxout. In this paper, we introduce a simple, new type of activation function for multilayer feed-forward architectures. Unlike other approaches where new activation functions have been designed by discarding many of the mainstays of traditional activation function design, our proposed function relies on them and therefore shares most of the properties found in traditional activation functions. Nevertheless, our activation function differs from traditional activation functions on two major points: its asymptote and global extremum. Defining a function which enjoys the property of having a global maximum and minimum, turned out to be critical during our design-process since we believe it is one of the main reasons behind the gap observed in performance between traditional activation functions and their recently introduced counterparts. We evaluate the effectiveness of the proposed activation function on four commonly used datasets, namely, MNIST, CIFAR-10, CIFAR-100, and the Pang and Lee’s movie review. Experimental results demonstrate that the proposed function can effectively be applied across various datasets where our accuracy, given the same network topology, is competitive with the state-of-the-art. In particular, the proposed activation function outperforms the state-of-the-art methods on the MNIST dataset.     

稀疏信号重构的罚函数神经网络模型

在压缩感知理论中,针对未知信号的稀疏性和信号非零元素位置的不确定性使得稀疏信号的重构比较困难,以及基于贪婪迭代方法的匹配追踪算法和基于凸松弛方法的基追踪算法对稀疏信号的重构概率不高的问题,提出一个罚函数神经网络模型.首先在感知矩阵满足有限等距性(RIP)的前提下,压缩感知问题可以转化为等价的l1-范数最小化问题.然后基...     

Output value-based initialization for radial basis function neural networks

The use of Radial Basis Function Neural Networks (RBFNNs) to solve functional approximation problems has been addressed many times in the literature. When designing an RBFNN to approximate a function, the first step consists of the initialization of the centers of the RBFs. This initialization task is very important because the rest of the steps are based on the positions of the centers. Many clustering techniques have been applied for this purpose achieving good results although they were constrained to the clustering problem. The next step of the design of an RBFNN, which is also very important, is the initialization of the radii for each RBF. There are few heuristics that are used for this problem and none of them use the information provided by the output of the function, but only the centers or the input vectors positions are considered. In this paper, a new algorithm to initialize the centers and the radii of an RBFNN is proposed. This algorithm uses the perspective of activation grades for each neuron, placing the centers according to the output of the target function. The radii are initialized using the center’s positions and their activation grades so the calculation of the radii also uses the information provided by the output of the target function. As the experiments show, the performance of the new algorithm outperforms other algorithms previously used for this problem.     

Effective incorporation of function into logic

During the last decade, the integration of functional and logic languages has received widespread attraction for the purpose of offering two different programming styles in one system simultaneously. The main goal is to incorporate the characteristics of the two paradigms coherently, without degrading the performance of the whole system. However, few languages have achieved this goal. Some of them, even though they have a rich set of functions, perform poorly. Others are efficient, but lose some important facilities. The paper proposes a functional logic language Lazy Aflog and its abstract machine FWAM-II as an expressive and efficient mechanism for this incorporation. Lazy Aflog is an extension of logic language in which functions are reduced in the extended unification, called E-unification with lazy evaluation. This extended unification allows Lazy Aflog to process infinite data structures and higher-order functions naturally. FWAM-II is an extension of the Warren Abstract Machine (WAM) in which the instructions and run-time structures to provide the suspension/reactivation of functional closure are added. These facilities enable FWAM-II to support not only resolution but also infinite data structures and higher-order functions efficiently. In addition, the experimental results show that Lazy Aflog and FWAM-II could be a good compromise between expressiveness and efficiency of the integration.     

Effective diagnosis of heart disease through neural networks ensembles

In the last decades, several tools and various methodologies have been proposed by the researchers for developing effective medical decision support systems. Moreover, new methodologies and new tools are continued to develop and represent day by day. Diagnosing of the heart disease is one of the important issue and many researchers investigated to develop intelligent medical decision support systems to improve the ability of the physicians. In this paper, we introduce a methodology which uses SAS base software 9.1.3 for diagnosing of the heart disease. A neural networks ensemble method is in the centre of the proposed system. This ensemble based methods creates new models by combining the posterior probabilities or the predicted values from multiple predecessor models. So, more effective models can be created. We performed experiments with the proposed tool. We obtained 89.01% classification accuracy from the experiments made on the data taken from Cleveland heart disease database. We also obtained 80.95% and 95.91% sensitivity and specificity values, respectively, in heart disease diagnosis.     

Design of a radial basis function neural network with a radius-modification algorithm using response surface methodology

A radial basis function (RBF) neural network was designed for time series forecasting using both an adaptive learning algorithm and response surface methodology (RSM). To improve the traditional RBF networks forecasting capability, the generalized delta rule learning method was employed to modify the radius of the kernel function. Then RSM was utilized to explore the mean square error response surface so that the appropriate combination of network parameters, such as the number of hidden nodes and the initial learning rates, could be found. Extensive studies were performed on the effect of the initial values of connection weights on the accuracy of the backpropagation learning method that was employed in the training of the RBF artificial neural network. The effectiveness of the neural network with the proposed radius-modification technique and the RSM method was demonstrated with an example of forecasting intensity pulsations of a laser. It was found that, by utilizing the proposed techniques, the neural network provided a more accurate prediction of the response.