Adjustable entropy function method for support vector machine

Based on KKT complementary condition in optimization theory, an unconstrained non-differential optimization model for support vector machine is proposed. An adjustable entropy function method is given to deal with the proposed optimization problem and the Newton algorithm is used to figure out the optimal solution. The proposed method can find an optimal solution with a relatively small parameter p, which avoids the numerical overflow in the traditional entropy function methods. It is a new approach to solve support vector machine. The theoretical analysis and experimental results illustrate the feasibility and efficiency of the proposed algorithm.

ENTROPY FUNCTIONAL FOR CONTINUOUS SYSTEMS OF FINITE ENTROPY

In this article, we introduce the concept of entropy functional for continuous systems on compact metric spaces, and prove some of its properties. We also extract the Kolmogorov entropy from the entropy functional.

Properties of Fuzzy Entropy Based on the Shape Change of Membership Function

Modification of a fuzzy partition often leads to the change of fuzziness of a fuzzy system. Researches on the change of fuzzy entropy of a fuzzy set, responding to shape alteration of membership function, therefore, play a significant role in analysis of the change of fuzziness of a fuzzy system because a fuzzy partition consists of a set of fuzzy sets which satisfy some special constraints. This paper has shown several results about entropy changes of a fuzzy set. First, the entropies of two same type of fuzzy sets have a constant proportional relationship which depends on the ratio of the sizes of their support intervals. Second, as for Triangular Fuzzy Numbers （TFNs）, the entropies of any two TFNs which can not be always the same type, also, have a constant proportional relationship which depends on the ratio of the sizes of their support intervals. Hence, any two TFNs with the same sizes of support intervals have the same entropies. Third, concerning two Triangular Fuzzy Sets （TFSs） with same sizes of support intervals and different heights, the relationship of their entropies lies on their height. Finally, we point it out a mistake that Chen＇s assertion that the entropy of resultant fuzzy set of elevation operation is directly to that of the original one while elevation factor just acts as a propartional factor. These results should contribute to the analysis and design of a fuzzy system.

Entropy Function-Based Algorithms for Solving a Class of Nonconvex Minimization Problems

Recently,the l_(p)minimization problem(p∈(0,1))for sparse signal recovery has been studied a lot because of its efficiency.In this paper,we propose a general smoothing algorithmic framework based on the entropy function for solving a class of l_(p)minimization problems,which includes the well-known unconstrained l_(2)-l_(p)problem as a special case.We show that any accumulation point of the sequence generated by the proposed algorithm is a stationary point of the l_(p)minimization problem,and derive a lower bound for the nonzero entries of the stationary point of the smoothing problem.We implement a specific version of the proposed algorithm which indicates that the entropy function-based algorithm is effective.

INTERVAL ADJUSTABLE ENTROPY ALGORITHM FOR A CLASS OF UNCONSTRAINED DISCRETE MINIMAX PROBLEMS

In this paper,a class of unconstrained discrete minimax problems is described,in which the objective functions are in C 1.The paper deals with this problem by means of taking the place of maximum entropy function with adjustable entropy function.By constructing an interval extension of adjustable entropy function an d some region deletion test rules,a new interval algorithm is presented.The rele vant properties are proven.The minimax value and the localization of the minimax points of the problem can be obtained by this method. This method can overcome the flow problem in the maximum entropy algorithm.Both theoretical and numerica l results show that the method is reliable and efficient.

An Interval Maximum Entropy Method for Quadratic Programming Problem

With the idea of maximum entropy function and penalty function methods, we transform the quadratic programming problem into an unconstrained differentiable optimization problem, discuss the interval extension of the maximum entropy function, provide the region deletion test rules and design an interval maximum entropy algorithm for quadratic programming problem. The convergence of the method is proved and numerical results are presented. Both theoretical and numerical results show that the method is reliable and efficient.

Adjustable entropy method for solving convex inequality problem

To solve the inequality problem, an adjustable entropy method is proposed. An inequality problem can be transformed into a minimax problem which is nondifferentiable; then an adjustable entropy is used to smooth the minimax problem. The solution of inequalities can be approached by using a BFGS algorithm of the standard optimization method. Some properties of the new approximate function are presented and then the global convergence are given according to the algorithm. Two numerical examples illustrate that the proposed method is efficient and is superior to the former ones.

Radar emitter signal recognition method based on improved collaborative semi-supervised learning

Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recogni-tion.To solve this problem,an optimized cooperative semi-supervised learning radar emitter recognition method based on a small amount of labeled data is developed.First,a small amount of labeled data are randomly sampled by using the bootstrap method,loss functions for three common deep learning net-works are improved,the uniform distribution and cross-entropy function are combined to reduce the overconfidence of softmax classification.Subsequently,the dataset obtained after sam-pling is adopted to train three improved networks so as to build the initial model.In addition,the unlabeled data are preliminarily screened through dynamic time warping(DTW)and then input into the initial model trained previously for judgment.If the judg-ment results of two or more networks are consistent,the unla-beled data are labeled and put into the labeled data set.Lastly,the three network models are input into the labeled dataset for training,and the final model is built.As revealed by the simula-tion results,the semi-supervised learning method adopted in this paper is capable of exploiting a small amount of labeled data and basically achieving the accuracy of labeled data recognition.

Global existence of weak solutions to a prey-predator model with strong cross-diffusion

Using finite differences and entropy inequalities, the global existence of weak solutions to a multidimensional parabolic strongly coupled prey-predator model is obtained. The nonnegativity of the solutions is also shown.

Improved response surface method and its application in stability reliability degree analysis of tunnel surrounding rock

An approach of limit state equation for surrounding rock was put forward based on deformation criterion. A method of symmetrical sampling of basic random variables adopted by classical response surface method was mended, and peak value and deflection degree of basic random variables distribution curve were took into account in the mended sampling method. A calculation way of probability moment, based on mended Rosenbluth method, suitable for non-explicit performance function was put forward. The first, second, third and fourth order moments of functional function value were calculated by mended Rosenbluth method through the first, second, third and fourth order moments of basic random variable. A probability density the function（PDF） of functional function was deduced through its first, second, third and fourth moments, the PDF in the new method took the place of the method of quadratic polynomial to approximate real functional function and reliability probability was calculated through integral by the PDF for random variable of functional function value in the new method. The result shows that the improved response surface method can adapt to various statistic distribution types of basic random variables, its calculation process is legible and need not itemtive circulation. In addition, a stability probability of surrounding rock for a tunnel was calculated by the improved method, whose workload is only 30% of classical method and its accuracy is comparative.

Estimating Weibull Parameters Using Least Squares and Multilayer Perceptron vs. Bayes Estimation

The Weibull distribution is regarded as among the finest in the family of failure distributions.One of the most commonly used parameters of the Weibull distribution(WD)is the ordinary least squares(OLS)technique,which is useful in reliability and lifetime modeling.In this study,we propose an approach based on the ordinary least squares and the multilayer perceptron(MLP)neural network called the OLSMLP that is based on the resilience of the OLS method.The MLP solves the problem of heteroscedasticity that distorts the estimation of the parameters of the WD due to the presence of outliers,and eases the difficulty of determining weights in case of the weighted least square(WLS).Another method is proposed by incorporating a weight into the general entropy(GE)loss function to estimate the parameters of the WD to obtain a modified loss function(WGE).Furthermore,a Monte Carlo simulation is performed to examine the performance of the proposed OLSMLP method in comparison with approximate Bayesian estimation(BLWGE)by using a weighted GE loss function.The results of the simulation showed that the two proposed methods produced good estimates even for small sample sizes.In addition,the techniques proposed here are typically the preferred options when estimating parameters compared with other available methods,in terms of the mean squared error and requirements related to time.

Vibration Fatigue Probabilistic Life Prediction Model and Method for Blade

Vibration fatigue is one of the main failure modes of blade.The vibration fatigue life of blade is scattered caused by manufacture error,material property dispersion and external excitation randomness.A new vibration fatigue probabilistic life prediction model(VFPLPM)and a prediction method are proposed in this paper.Firstly,as one-dimensional volumetric method(ODVM)only considers the principle calculation direction,a three-dimensional space vector volumetric method(TSVVM)is proposed to improve fatigue life prediction accuracy for actual threedimensional engineering structure.Secondly,based on the two volumetric methods(ODVM and TSVVM),the material C-P-S-N fatigue curve model(CFCM)and the maximum entropy quantile function model(MEQFM),VFPLPM is established to predict the vibration fatigue probabilistic life of blade.The VFPLPM is combined with maximum stress method(MSM),ODVM and TSVVM to estimate vibration fatigue probabilistic life of blade simulator by finite element simulation,and is verified by vibration fatigue test.The results show that all of the three methods can predict the vibration fatigue probabilistic life of blade simulator well.VFPLPM &TSVVM method has the highest computational accuracy for considering stress gradient effect not only in the principle calculation direction but also in other space vector directions.

3D laser scanning strategy based on cascaded deep neural network

A 3D laser scanning strategy based on cascaded deep neural network is proposed for the scanning system converted from 2D Lidar with a pitching motion device. The strategy is aimed at moving target detection and monitoring. Combining the device characteristics, the strategy first proposes a cascaded deep neural network, which inputs 2D point cloud, color image and pitching angle. The outputs are target distance and speed classification. And the cross-entropy loss function of network is modified by using focal loss and uniform distribution to improve the recognition accuracy. Then a pitching range and speed model are proposed to determine pitching motion parameters. Finally, the adaptive scanning is realized by integral separate speed PID. The experimental results show that the accuracies of the improved network target detection box, distance and speed classification are 90.17%, 96.87% and 96.97%, respectively. The average speed error of the improved PID is 0.4239°/s, and the average strategy execution time is 0.1521 s.The range and speed model can effectively reduce the collection of useless information and the deformation of the target point cloud. Conclusively, the experimental of overall scanning strategy show that it can improve target point cloud integrity and density while ensuring the capture of target.

Thermodynamic properties of ZnSe under pressure and with variation in temperature

The thermodynamic properties of Zn Se are obtained by using quasi-harmonic Debye model embedded in Gibbscode for pressure range 0–10 GPa and for temperature range 0–1000 K. Helmholtz free energy, internal energy, entropy,Debye temperature, and specific heat are calculated. The thermal expansion coefficient along with Gruneisen parameter are also calculated at room temperature for the pressure range. It is found that internal energy is pressure dependent at low temperature, whereas entropy and Helmholtz free energy are pressure sensitive at high temperature. At ambient conditions,the obtained results are found to be in close agreement to available theoretical and experimental data.

The altered network complexity of resting-state functional brain activity in schizophrenia and bipolar disorder patients

Schizophrenia(SZ)and bipolar disorder(BD)are two of themost frequent mental disorders.These disorders exhibit similarpsychotic symptoms,making diagnosis challenging and leadingto misdiagnosis.Yet,the network complexity changes drivingspontaneous brain activity in SZ and BD patients are still unknown.Functional entropy(FE)is a novel way of measuring the dispersion(or spread)of functional connectivities inside the brain.The FE wasutilized in this study to examine the network complexity of the resting-state fMRI data of SZ and BD patients at three levels,including global,modules,and nodes.At three levels,the FE of SZand BD patients was considerably lower than that of normal control(NC).At the intra-module level,the FE of SZ was substantially higher than that of BD in the cingulo-opercular network.Moreover,a strong negative association between FE and clinical measureswas discovered in patient groups.Finally,we classified using theFE features and attained an accuracy of 66.7%(BD vs.SZ vs.NC)and an accuracy of 75.0%(SZ vs.BD).These findings proposed that network connectivity’s complexity analyses using FE can provideimportant insights for the diagnosis of mental illness.

Maximum-entropy clustering algorithm and its global convergence analysis

Constructing a batch of differentiable entropy functions touniformly approximate an objective function by means of the maximum-entropy principle, a new clustering algorithm, called maximum-entropy clustering algorithm, is proposed based on optimization theory. This algorithm is a soft generalization of the hard C-means algorithm and possesses global convergence. Its relations with other clustering algorithms are discussed.

The Symbolic Extension Theory in Topological Dynamics

In this survey we will present the symbolic extension theory in topological dynamics,which was built over the past twenty years.

Reliability analysis based on a novel density estimation method for structures with correlations

Estimating the Probability Density Function（PDF） of the performance function is a direct way for structural reliability analysis,and the failure probability can be easily obtained by integration in the failure domain.However,efficiently estimating the PDF is still an urgent problem to be solved.The existing fractional moment based maximum entropy has provided a very advanced method for the PDF estimation,whereas the main shortcoming is that it limits the application of the reliability analysis method only to structures with independent inputs.While in fact,structures with correlated inputs always exist in engineering,thus this paper improves the maximum entropy method,and applies the Unscented Transformation（UT） technique to compute the fractional moments of the performance function for structures with correlations,which is a very efficient moment estimation method for models with any inputs.The proposed method can precisely estimate the probability distributions of performance functions for structures with correlations.Besides,the number of function evaluations of the proposed method in reliability analysis,which is determined by UT,is really small.Several examples are employed to illustrate the accuracy and advantages of the proposed method.

Statistical Inference of Exponentiated Moment Exponential Distribution Based on Lower Record Values

Based on lower record values,we first derive the exact explicit expressions as well as recurrence relations for the single and product moments of record values and then use these results to compute the means,variances and coefficient of skewness and kurtosis of exponentiated moment exponential distribution(EMED),a new extension of moment exponential distribution,recently introduced by Hasnain(Exponentiated moment exponential distribution.Ph.D.Thesis,2013).Next we obtain the maximum likelihood estimators of the unknown parameters and the approximate confidence intervals of the EMED.Finally,we consider Bayes estimation under the symmetric and asymmetric loss functions using gamma priors for both shape and scale parameters.We have also derived the Bayes interval of this distribution and discussed both frequentist and the Bayesian prediction intervals of the future record values based on the observed record values.Monte Carlo simulations are performed to compare the performances of the proposed methods,and a data set has been analyzed for illustrative purposes.