The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
This paper is to prove a widely believed '1-bit-matching' conjecture in independent component analysis (ICA) community, which is stated as "all the sources can be separated as long as there is 1-to-1 same-sign-correspondence between the kurtosis signs of all source probability density functions (pdf's) and the kurtosis signs of all model pdf's"
This paper proposes a novel immune clonal algorithm, called a quantum-inspired immune clonal algorithm (QICA), which is based on the concept and principles of quantum computing, such as a quantum bit and superposition of states. Like other evolutionary algorithms, QICA is also characterized by the representation of the individual, the evaluation function, and the population dynamics. However, instead...
Drosophila flies can be trained in the flight simulator to operantly avoid heat by choosing certain flight orientations relative to landmarks. Flies primarily store pattern orientations associated with the absence of heat. They readily escape from heat-associated orientations under the direct influence of the reinforcer but not in the subsequent memory tests. This paper shows that Drosophila flies...
In this paper, an intelligent system is featured by both its abilities of interpreting what are observed via discovering knowledge about the world it survives, and its problem solving skills of handling each issue encountered in the world. Correspondingly, the abilities and skills are obtained by two types of learning via evidences or data from the world. Due to noises in observation and a finite...
We find very tight bounds on the accuracy of a support vector machine classification error within the algorithmic inference framework. The framework is specially suitable for this kind of classifier since (i) we know the number of support vectors really employed, as an ancillary output of the learning procedure, and (ii) we can appreciate confidence intervals of misclassifying probability exactly...
In radar system the anti-jamming performance has always been a problem for a long time. Radar can acquire a better detecting and tracking ability if the problem is solved. The limitations of traditional signal processing methods have become more evident under complicated environments. In this paper, we analyze the flow of radar signal processing and the methods of canonical correlation analysis (CCA)...
This paper proposes a novel method of combining rough concepts with neural computation. The proposed rough neuron consists of, one lower bound neuron and another boundary neuron. The combination is designed in such a way that the boundary neuron deals only with the random and unpredictable part of the applied signal. Such architecture effectively prunes the search space for the respective constituent...
Support vector classification with Gaussian RBF kernel is sensitive to the kernel width. Small kernel width may cause over-fitting, and large one under-fitting. The so-called optimal kernel width is merely selected based on the tradeoff between under-fitting loss and over-fitting loss. So, there exists urgent need to further reduce the tradeoff loss. To circumvent this, we scale the kernel width in...
In this work, a neural network material model is built for the simulation of the inelastic behavior of biocomposite insect cuticle. Radial basis function neural network is adopted in the simulation for that the neural network has the characteristic of fast and exactly completing the simulation. In the construction of the neural network, the network is trained based on the experimental data of the...
Cascade-correlation algorithm has usually been used for training feedforward multilayer networks with sigmoidal activation in the hidden units. Modeling using modified cascade-correlation radial basis function (CCRBF) networks for an experimental platform is presented. The behavior of the four degrees-of-freedom (DOF) platform exemplifies a high order nonlinear system with significant cross-coupling...
A novel method using artificial neural network with back-propagation algorithm to detect and separate LFM signal is proposed. This method trains the network by LFM signal mixed with Gauss noise. Simulation result shows the trained BP neural network can eliminate noise effectively. In addition, if the learning sample is a multicomponent LFM signal, the trained network can separate the LFM signal component...
Recent studies on generalized congruence neural network show that such network has fast convergence rate. But the reason is unknown. In this paper, through the analysis on the error function of the generalized congruence neuron, we find that the fast convergence is due to the multiple minima, which is generated by the generalized congruence activation function. This conclusion is verified by two experiments
A combination classification algorithm, ER-SVM, is proposed to improve the generalization performance of support vector machine (SVM) by directly making full use of the empirical risk (ER) information of SVM in the paper. SVM classification is the implementation of structure risk minimization (SRM) principle. SVM may achieve SRM from the minimal summation of ER and VC confidence according to the theory...
A genetic algorithm simulating evolution is proposed to yield near optional solution to the traveling salesman problem. Noting that Darwinian evolution is itself optimization process, we propose a heuristic algorithm that incorporates the tents of natural selection. The time complexity of this algorithm is equivalent to the fastest sorting scheme. The algorithm is used to solve the China - traveling...
In this paper, a new dynamical multi-objective evolutionary algorithm based on the information entropy is proposed inspired by the principle of minimal free energy from the statistical mechanics. Developed to solving multi-objective optimization problems, the maintenance of the diversity of the population is essentially considered in this new algorithm by using the information entropy. The numerical...
Based on fuzzy association degree, a new pattern recognition algorithm is set up. First, some new concepts of fuzzy association coefficient (FAC), fuzzy association degree (FAD) and fuzzy relative weight (FRW) have been proposed for surveying data information. Second, on the basis of the concepts proposed here, a new pattern recognition algorithm has been set up. At last, the algorithm set up here...
According to the shortages of application of MRS and MRI to the clinical cancer diagnosis, an effective method to analyze and process the raw data of nuclear magnetic resonance is brought forward based on wavelet transform and pattern recognition technologies. Aiming at the characteristics of FID signals and MRS, de-nosing of FID and MRS data was performed using wavelet threshold to obtain the better...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.