Search

Visual recognition of crowd dynamics has had a huge impact on several applications including surveillance, situation awareness, homeland security and intelligent environments. However, the state-of-the-art in crowd analysis has become diverse due to factors such as: (a) the underlying definition of a crowd, (b) the constituent elements of the crowd processing model, (c) its application, hence (d) the dataset and (e) the evaluation criteria available for benchmarking. Although such diversity is healthy, the techniques for crowd modelling thus developed have failed to establish credibility therefore becoming unreliable and of questionable validity across different research disciplines. This paper aims to give an account of such issues by deducing key statistical evidence from the existing literature and providing recommendations towards focusing on the general aspects of techniques rather than any specific algorithm. The advances and trends in crowd analysis are drawn in the context of crowd modelling studies published in leading journals and conferences over the past 7 years. Finally, this paper shall also provide a qualitative and quantitative comparison of some existing methods using various publicly available crowd datasets to substantiate some of the theoretical claims. [ABSTRACT FROM AUTHOR]

Linear Discriminant Analysis (LDA) is a standard tool for classification and dimension reduction in many applications. However, the problem of high dimension is still a great challenge for the classical LDA. In this paper we consider the supervised pattern classification in the high dimensional setting, in which the number of features is much larger than the number of observations and present a novel approach to the sparse optimal scoring problem using the zero-norm. The difficulty in treating the zero-norm is overcome by using appropriate continuous approximations such that the resulting problems are solved by alternating schemes based on DC (Difference of Convex functions) programming and DCA (DC Algorithms). The experimental results on both simulated and real datasets show the efficiency of the proposed algorithms compared to the five state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Nonlinear optimal control (NOC) problem with complex dynamic constraints (CDC) is difficult to compute even with direct method. In this paper, a hybrid two-stage approach integrating an improved backtracking search optimization algorithm (IBSA) with the hp-adaptive Gauss pseudo-spectral methods (hpGPM) is proposed. Firstly, BSA is improved to enhance its convergent speed and the global search ability, by adopting the harmony search strategy and an adaptive amplitude control factor with individual optimum fitness feedback. Then, at the beginning stage of the hybrid search process, an initialization generator is constructed using IBSA to find a near optimum solution. When the change in fitness function approaches to a predefined value which is small enough, the search process is replaced by hpGPM to accelerate the search process and find an accurate solution. By this way, the hybrid algorithm is able to find a global optimum more quickly and accurately. Two NOC problems with CDC are examined using the proposed algorithm, and the corresponding Monte Carlo simulations are conducted. The comparison results show the hybrid algorithm achieves better performance in convergent speed, accuracy and robustness. [ABSTRACT FROM AUTHOR]

Minimum component analysis is an elegant and promising approach to system identification. This paper focuses on developing related recursive algorithms. Two algorithms are presented. In the first approach, rank-one modification is used to get an accurate and quickly convergent algorithm. However, its drawback is low computational efficiency. Therefore, the second algorithm is proposed to realize faster computation. The simulation results verify the effectiveness of both algorithms. [ABSTRACT FROM AUTHOR]

Dynamic model is an important issue for research on stability, dynamic characteristics, surge and control technique of axial-flow pumping system, and such a model is usually characterized by complex nonlinearity, strong coupling and time-varying mathematical equation. For the convenience of establishing model and highly effective computing, dynamic characteristics of the whole system are divided into four parts: pump lift-flow characteristics, pipeline characteristics, mechanical characteristics of asynchronous motor and torque characteristics of pump load. Each part is a nonlinear subsystem, and there are complex coupling relations among each other. In the paper, each part of the pump system is modeled respectively by mechanisms of hydrodynamics, transmission dynamics, electromechanics and affinity law. Considering that the axial-flow pump is characterized by nonlinearity and parameters are difficultly estimated in the low flow operation area and that the data of pump head-flow can be easily tested under the speed of power frequency, a modeling method combined with the RBF neural network is proposed, where hidden layer parameters are optimized by K means clustering algorithm, and the weights are trained by least square method. At last the whole simulation model of the axial-flow pumping system is set up, and the validity of the proposed modeling method is verified through simulation. [ABSTRACT FROM AUTHOR]

A unified associative memory model with a novel method for designing associative memories is presented in this paper. Based on continuous recurrent neural networks, bipolar patterns inputted from external can cause the output of neural networks to be memorized patterns. In the method, two conditions relevant to external inputs are derived to ensure the network states converge to a stable interval, and an exponential stable criterion is proposed for the network being a bipolar associative memory with higher recall speed. By introducing a tunable slope activation function and considering time delay, the proposed model is general and can recall the memorized patterns in auto-associative and hetero-associative way, while higher robust and more flexible memory can be obtained through the proposed method. Experimental verification demonstrates the effectiveness and generalization of the proposed method. [ABSTRACT FROM AUTHOR]

The success of SVM in solving pattern recognition problems has encouraged researcher to extend the development of different versions. They are well-known for their robustness and good generalization performance. In many real-world applications, the data to be trained are available on-line in a sequential fashion and because of space and time requirements, batch training methods are not suitable. This paper proposes a new fast on-line algorithm called OTWISVM. It defines two optimization problems and incremental learning is done based of them. Two hyperplanes are generated as decision functions thus each of them is closer to one of the two classes and is as far as possible from the other. The solution is constructed via two subsets of linearly independent samples seen so far, and is always bounded. Good accuracy and notable speed of the method was tested and affirmed both on ordinary and noisy data sets as opposed to similar algorithms. [ABSTRACT FROM AUTHOR]

This paper proposes an improved stochastic second order learning algorithm for supervised neural network training. The proposed algorithm, named bounded stochastic diagonal Levenberg–Marquardt (B-SDLM), utilizes both gradient and curvature information to achieve fast convergence while requiring only minimal computational overhead than the stochastic gradient descent (SGD) method. B-SDLM has only a single hyperparameter as opposed to most other learning algorithms that suffer from the hyperparameter overfitting problem due to having more hyperparameters to be tuned. Experiments using the multilayer perceptron (MLP) and convolutional neural network (CNN) models have shown that B-SDLM outperforms other learning algorithms with regard to the classification accuracies and computational efficiency (about 5.3% faster than SGD on the mnist-rot-bg-img database). It can classify all testing samples correctly on the face recognition case study based on AR Purdue database. In addition, experiments on handwritten digit classification case studies show that significant improvements of 19.6% on MNIST database and 17.5% on mnist-rot-bg-img database can be achieved in terms of the testing misclassification error rates (MCRs). The computationally expensive Hessian calculations are kept to a minimum by using just 0.05% of the training samples in its estimation or updating the learning rates once per two training epochs, while maintaining or even achieving lower testing MCRs. It is also shown that B-SDLM works well in the mini-batch learning mode, and we are able to achieve 3.32 × performance speedup when deploying the proposed algorithm in a distributed learning environment with a quad-core processor. [ABSTRACT FROM AUTHOR]

In this paper, we consider the quantized distributed optimization problem over general directed digital multi-agent networks, where the communication channels have limited data transmission rates. To solve the optimization problem, a distributed quantized subgradient algorithm is presented among agents. Based on an encoder–decoder scheme and a zoom-in technique, we can achieve not only a consensus, but also an optimal solution. In particular, we study two cases of the quantization levels of each connected directed digital communication channel. One is under the case that the quantization levels are time-varying at each time step, and the other is under the case of fixed quantization level. Two rigorous theoretical analyses are performed and the optimal solutions can be obtained asymptotically. Moreover, the upper bound of the quantization levels at each time step and the convergence rate are analytically characterized. The effectiveness of proposed algorithm is demonstrated by two illustrative examples. [ABSTRACT FROM AUTHOR]

Abstract: Recently, Gandomi and Alavi proposed a meta-heuristic optimization algorithm, called Krill Herd (KH), for global optimization [Gandomi AH, Alavi AH. Krill Herd: A New Bio-Inspired Optimization Algorithm. Communications in Nonlinear Science and Numerical Simulation, 17(12), 4831–4845, 2012.]. This paper represents an optimization method to global optimization using a novel variant of KH. This method is called the Stud Krill Herd (SKH). Similar to genetic reproduction mechanisms added to KH method, an updated genetic reproduction schemes, called stud selection and crossover (SSC) operator, is introduced into the KH during the krill updating process dealing with numerical optimization problems. The introduced SSC operator is originated from original Stud genetic algorithm. In SSC operator, the best krill, the Stud, provides its optimal information for all the other individuals in the population using general genetic operators instead of stochastic selection. This approach appears to be well capable of solving various functions. Several problems are used to test the SKH method. In addition, the influence of the different crossover types on convergence and performance is carefully studied. Experimental results indicate an instructive addition to the portfolio of swarm intelligence techniques. [Copyright &y& Elsevier]

Abstract: Hybrid strategy, which generalizes a specific single-label algorithm while one or two data decomposition tricks are applied implicitly or explicitly, has become an effective and efficient tool to design and implement various multi-label classification algorithms. In this paper, we extend traditional binary support vector machine by introducing an approximate ranking loss as its empirical loss term to build a novel support vector machine for multi-label classification, resulting into a quadratic programming problem with different upper bounds of variables to characterize label correlation of individual instance. Further, our optimization problem can be solved via combining one-versus-rest data decomposition trick with modified binary support vector machine, which dramatically reduces computational cost. Experimental study on ten multi-label data sets illustrates that our method is a powerful candidate for multi-label classification, compared with four state-of-the-art multi-label classification approaches. [Copyright &y& Elsevier]

Abstract: In this paper, one novel genetic algorithm dynamic chain-like agent genetic algorithm (CAGA) is proposed for solving global numerical optimization problem and feature selection problem. The CAGA combines the chain-like agent structure with dynamic neighboring genetic operators to get higher optimization capability. An agent in chain-like agent structure represents a candidate solution to the optimization problem. Any agent interacts with neighboring agents to evolve. With dynamic neighboring genetic operators, they compete and cooperate with their neighbors, and can use knowledge to increase energies. Global numerical optimization problem and feature selection problem are the most important problems for evolutionary algorithm, especially for genetic algorithm. Hence, the experiments of global numerical optimization and feature selection are necessary to verify the performance of genetic algorithms. Corresponding experiments have been done and show that CAGA is suitable for real coding and binary coding optimization problems, and has more precise and more stable optimization results. [Copyright &y& Elsevier]

In this paper we present a method of parameter optimization, relative trust-region learning, where the trust-region method and the relative optimization [M. Zibulevsky, Blind source separation with relative Newton method, in: Proceedings of the ICA, Nara, Japan, 2003, pp. 897–902] are jointly exploited. The relative trust-region method finds a direction and a step size with the help of a quadratic model of the objective function (as in the conventional trust-region methods) and updates parameters in a multiplicative fashion (as in the relative optimization). We apply this relative trust-region learning method to the problem of independent component analysis (ICA), which leads to the relative TR-ICA algorithm which turns out to possess the equivariant property (as in the relative gradient) and to achieve faster convergence than the relative gradient and even Newton-type algorithms. Empirical comparisons with several existing ICA algorithms demonstrate the useful behavior of the relative TR-ICA algorithm, such as the equivariant property and fast convergence. [Copyright &y& Elsevier]