Cross entropy is a measure in machine learning and deep learning that assesses the difference between predicted and actual probability distributions. In this study, we propose cross entropy as a performance evaluation...Cross entropy is a measure in machine learning and deep learning that assesses the difference between predicted and actual probability distributions. In this study, we propose cross entropy as a performance evaluation metric for image classifier models and apply it to the CT image classification of lung cancer. A convolutional neural network is employed as the deep neural network (DNN) image classifier, with the residual network (ResNet) 50 chosen as the DNN archi-tecture. The image data used comprise a lung CT image set. Two classification models are built from datasets with varying amounts of data, and lung cancer is categorized into four classes using 10-fold cross-validation. Furthermore, we employ t-distributed stochastic neighbor embedding to visually explain the data distribution after classification. Experimental results demonstrate that cross en-tropy is a highly useful metric for evaluating the reliability of image classifier models. It is noted that for a more comprehensive evaluation of model perfor-mance, combining with other evaluation metrics is considered essential. .展开更多
This study evaluates the performance and reliability of a vision transformer (ViT) compared to convolutional neural networks (CNNs) using the ResNet50 model in classifying lung cancer from CT images into four categori...This study evaluates the performance and reliability of a vision transformer (ViT) compared to convolutional neural networks (CNNs) using the ResNet50 model in classifying lung cancer from CT images into four categories: lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), large cell carcinoma (LULC), and normal. Although CNNs have made significant advancements in medical imaging, their limited capacity to capture long-range dependencies has led to the exploration of ViTs, which leverage self-attention mechanisms for a more comprehensive global understanding of images. The study utilized a dataset of 748 lung CT images to train both models with standardized input sizes, assessing their performance through conventional metrics—accuracy, precision, recall, F1 score, specificity, and AUC—as well as cross entropy, a novel metric for evaluating prediction uncertainty. Both models achieved similar accuracy rates (95%), with ViT demonstrating a slight edge over ResNet50 in precision and F1 scores for specific classes. However, ResNet50 exhibited higher recall for LULC, indicating fewer missed cases. Cross entropy analysis showed that the ViT model had lower average uncertainty, particularly in the LUAD, Normal, and LUSC classes, compared to ResNet50. This finding suggests that ViT predictions are generally more reliable, though ResNet50 performed better for LULC. The study underscores that accuracy alone is insufficient for model comparison, as cross entropy offers deeper insights into the reliability and confidence of model predictions. The results highlight the importance of incorporating cross entropy alongside traditional metrics for a more comprehensive evaluation of deep learning models in medical image classification, providing a nuanced understanding of their performance and reliability. While the ViT outperformed the CNN-based ResNet50 in lung cancer classification based on cross-entropy values, the performance differences were minor and may not hold clinical significance. Therefore, it may be premature to consider replacing CNNs with ViTs in this specific application.展开更多
In order to increase productivity and reduce energy consumption of steelmaking-continuous casting(SCC) production process, especially with complicated technological routes, the cross entropy(CE) method was adopted to ...In order to increase productivity and reduce energy consumption of steelmaking-continuous casting(SCC) production process, especially with complicated technological routes, the cross entropy(CE) method was adopted to optimize the SCC production scheduling(SCCPS) problem. Based on the CE method, a matrix encoding scheme was proposed and a backward decoding method was used to generate a reasonable schedule. To describe the distribution of the solution space, a probability distribution model was built and used to generate individuals. In addition, the probability updating mechanism of the probability distribution model was proposed which helps to find the optimal individual gradually. Because of the poor stability and premature convergence of the standard cross entropy(SCE) algorithm, the improved cross entropy(ICE) algorithm was proposed with the following improvements: individual generation mechanism combined with heuristic rules, retention mechanism of the optimal individual, local search mechanism and dynamic parameters of the algorithm. Simulation experiments validate that the CE method is effective in solving the SCCPS problem with complicated technological routes and the ICE algorithm proposed has superior performance to the SCE algorithm and the genetic algorithm(GA).展开更多
In order to improve the global search ability of biogeography-based optimization(BBO)algorithm in multi-threshold image segmentation,a multi-threshold image segmentation based on improved BBO algorithm is proposed.Whe...In order to improve the global search ability of biogeography-based optimization(BBO)algorithm in multi-threshold image segmentation,a multi-threshold image segmentation based on improved BBO algorithm is proposed.When using BBO algorithm to optimize threshold,firstly,the elitist selection operator is used to retain the optimal set of solutions.Secondly,a migration strategy based on fusion of good solution and pending solution is introduced to reduce premature convergence and invalid migration of traditional migration operations.Thirdly,to reduce the blindness of traditional mutation operations,a mutation operation through binary computation is created.Then,it is applied to the multi-threshold image segmentation of two-dimensional cross entropy.Finally,this method is used to segment the typical image and compared with two-dimensional multi-threshold segmentation based on particle swarm optimization algorithm and the two-dimensional multi-threshold image segmentation based on standard BBO algorithm.The experimental results show that the method has good convergence stability,it can effectively shorten the time of iteration,and the optimization performance is better than the standard BBO algorithm.展开更多
The segmentation effect of Tsallis entropy method is superior to that of Shannon entropy method, and the computation speed of two-dimensional Shannon cross entropy method can be further improved by optimization. The e...The segmentation effect of Tsallis entropy method is superior to that of Shannon entropy method, and the computation speed of two-dimensional Shannon cross entropy method can be further improved by optimization. The existing two-dimensional Tsallis cross entropy method is not the strict two-dimensional extension. Thus two new methods of image thresholding using two-dimensional Tsallis cross entropy based on either Chaotic Particle Swarm Optimization (CPSO) or decomposition are proposed. The former uses CPSO to find the optimal threshold. The recursive algorithm is adopted to avoid the repetitive computation of fitness function in iterative procedure. The computing speed is improved greatly. The latter converts the two-dimensional computation into two one-dimensional spaces, which makes the computational complexity further reduced from O(L2) to O(L). The experimental results show that, compared with the proposed recently two-dimensional Shannon or Tsallis cross entropy method, the two new methods can achieve superior segmentation results and reduce running time greatly.展开更多
A matrix encoding scheme for the steelmaking continuous casting( SCC) production scheduling( SCCPS) problem and the corresponding decoding method are proposed. Based on it,a cross entropy( CE) method is adopted and an...A matrix encoding scheme for the steelmaking continuous casting( SCC) production scheduling( SCCPS) problem and the corresponding decoding method are proposed. Based on it,a cross entropy( CE) method is adopted and an improved cross entropy( ICE) algorithm is proposed to solve the SCCPS problem to minimize total power consumption. To describe the distribution of the solution space of the CE method,a probability model is built and used to generate individuals by sampling and a probability updating mechanism is introduced to trace the promising samples. For the ICE algorithm,some samples are generated by the heuristic rules for the shortest makespan due to the relation between the makespan and the total power consumption,which can reduce the search space greatly. The optimal sample in each iteration is retained through a retention mechanism to ensure that the historical optimal sample is not lost so as to improve the efficiency and global convergence. A local search procedure is carried out on a part of better samples so as to improve the local exploitation capability of the ICE algorithm and get a better result. The parameter setting is investigated by the Taguchi method of design-of-experiment. A number of simulation experiments are implemented to validate the effectiveness of the ICE algorithm in solving the SCCPS problem and also the superiority of the ICE algorithm is verified through the comparison with the standard cross entropy( SCE) algorithm.展开更多
Emergence of drug resistant bacteria is one of the serious problems in today’s public health. However, the relationship between genomic mutation of bacteria and the phenotypic difference of them is still unclear. In ...Emergence of drug resistant bacteria is one of the serious problems in today’s public health. However, the relationship between genomic mutation of bacteria and the phenotypic difference of them is still unclear. In this paper, based on the mutation information in whole genome sequences of 96 MRSA strains, two kinds of phenotypes (pathogenicity and drug resistance) were learnt and predicted by machine learning algorithms. As a result of effective feature selection by cross entropy based sparse logistic regression, these phenotypes could be predicted in sufficiently high accuracy (100% and 97.87%, respectively) with less than 10 features. It means that we could develop a novel rapid test method in the future for checking MRSA phenotypes.展开更多
To consider multi-objective optimization problem with the number of feed array elements and sidelobe level of large antenna array, multi-objective cross entropy(CE) algorithm is proposed by combining fuzzy c-mean clus...To consider multi-objective optimization problem with the number of feed array elements and sidelobe level of large antenna array, multi-objective cross entropy(CE) algorithm is proposed by combining fuzzy c-mean clustering algorithm with traditional cross entropy algorithm, and specific program flow of the algorithm is given.Using the algorithm, large thinned array(200 elements) given sidelobe level(-10,-19 and-30 d B) problem is solved successfully. Compared with the traditional statistical algorithms, the optimization results of the algorithm validate that the number of feed array elements reduces by 51%, 11% and 6% respectively. In addition, compared with the particle swarm optimization(PSO) algorithm, the number of feed array elements from the algorithm is more similar, but the algorithm is more efficient.展开更多
A health status assessment method based on cross entropy and support vector machine(SVM)is proposed for the new urban rail vehicle traction systems.First,an index system for health assessment of the traction system is...A health status assessment method based on cross entropy and support vector machine(SVM)is proposed for the new urban rail vehicle traction systems.First,an index system for health assessment of the traction system is established,and combined weights of the index layer are obtained via cross entropy.Then,an SVM assessment model considering actual operating data and each status level of the traction system is established.Finally,the model is simulated in Matlab to obtain assessment results.The results indicate that the proposed method can provide the health status information of the traction system intuitively and complete the health status assessment of the traction system of the new urban rail vehicle effectively,by exploiting the traction system’s layered analysis model.The health status can be assessed accurately and reliably by adopting the cross entropy theory and SVM theory.展开更多
In the present study we have formulated a Minimum Cross Fuzzy Entropy Problem (Minx(F)EntP) and proposed sufficient conditions for existence of its solution. Mentioned problem can be formulated as follows. In the ...In the present study we have formulated a Minimum Cross Fuzzy Entropy Problem (Minx(F)EntP) and proposed sufficient conditions for existence of its solution. Mentioned problem can be formulated as follows. In the set of membership functions satisfying the given moment constraints generated by given moment functions it is required to choose the membership function that is closest to a priori membership function in the sense of cross fuzzy entropy measure. The existence of solution of formulated problem is proved by virtue of concavity property of cross fuzzy entropy measure, the implicit function theorem and Lagrange multipliers method. Moreover, Generalized Cross Fuzzy Entropy Optimization Methods in the form of MinMinx(F)EntM and MaxMinx(F)EntM are suggested on the basis of primary phase of minimizing cross fuzzy entropy measure for fixed moment vector function and on the definition of the special functional with Minx(F)Ent values of cross fuzzy entropy measure. Next phase for obtaining mentioned distributions consists of optimization of defined functional with respect to moment vector functions. Distributions obtained by mentioned methods are defined as (MinMinx(F)Ent)m and (MaxMinx(F)Ent)m distributions.展开更多
Value-at-Risk (VaR) estimation via Monte Carlo (MC) simulation is studied here. The variance reduction technique is proposed in order to speed up MC algorithm. The algorithm for estimating the probability of high ...Value-at-Risk (VaR) estimation via Monte Carlo (MC) simulation is studied here. The variance reduction technique is proposed in order to speed up MC algorithm. The algorithm for estimating the probability of high portfolio losses (more general risk measure) based on the Cross - Entropy importance sampling is developed. This algorithm can easily be applied in any light- or heavy-tailed case without an extra adaptation. Besides, it does not loose in the performance in comparison to other known methods. A numerical study in both cases is performed and the variance reduction rate is compared with other known methods. The problem of VaR estimation using procedures for estimating the probability of high portfolio losses is also discussed.展开更多
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...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.展开更多
This study propose a new robust method to rank the performances of multi-assets (portfolios), based purely on their return time series. This method makes no assumption on the distributions. Topsoe distance is symmet...This study propose a new robust method to rank the performances of multi-assets (portfolios), based purely on their return time series. This method makes no assumption on the distributions. Topsoe distance is symmetrized Kullback-Leibler divergence by average of the probabilities. The square root of Topsoe distance is a metric. We extend this metric from probability density functions to real number series on (0, 1 ]. We call it ST-metric. We show the consistency of ST-metric with mean-variance theory and stochastic dominance method of order one and two. We demonstrate the advantages of ST-metric over mean-variance rule and stochastic dominance method of order one and two.展开更多
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 monito...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.展开更多
The increasing penetration of highly intermittent wind generation could seriously jeopardize the operation reliability of power systems and increase the risk of electricity outages. To this end, this paper proposes a ...The increasing penetration of highly intermittent wind generation could seriously jeopardize the operation reliability of power systems and increase the risk of electricity outages. To this end, this paper proposes a novel data-driven method for operation risk assessment of wind-integrated power systems. Firstly, a new approach is presented to model the uncertainty of wind power in lead time. The proposed approach employs k-means clustering and mixture models(MMs) to construct time-dependent probability distributions of wind power.The proposed approach can also capture the complicated statistical features of wind power such as multimodality. Then, a nonsequential Monte Carlo simulation(NSMCS) technique is adopted to evaluate the operation risk indices. To improve the computation performance of NSMCS, a cross-entropy based importance sampling(CE-IS) technique is applied. The CE-IS technique is modified to include the proposed model of wind power.The method is validated on a modified IEEE 24-bus reliability test system(RTS) and a modified IEEE 3-area RTS while employing the historical data of wind generation. The simulation results verify the importance of accurate modeling of shortterm uncertainty of wind power for operation risk assessment.Further case studies have been performed to analyze the impact of transmission systems on operation risk indices. The computational performance of the framework is also examined.展开更多
文摘Cross entropy is a measure in machine learning and deep learning that assesses the difference between predicted and actual probability distributions. In this study, we propose cross entropy as a performance evaluation metric for image classifier models and apply it to the CT image classification of lung cancer. A convolutional neural network is employed as the deep neural network (DNN) image classifier, with the residual network (ResNet) 50 chosen as the DNN archi-tecture. The image data used comprise a lung CT image set. Two classification models are built from datasets with varying amounts of data, and lung cancer is categorized into four classes using 10-fold cross-validation. Furthermore, we employ t-distributed stochastic neighbor embedding to visually explain the data distribution after classification. Experimental results demonstrate that cross en-tropy is a highly useful metric for evaluating the reliability of image classifier models. It is noted that for a more comprehensive evaluation of model perfor-mance, combining with other evaluation metrics is considered essential. .
文摘This study evaluates the performance and reliability of a vision transformer (ViT) compared to convolutional neural networks (CNNs) using the ResNet50 model in classifying lung cancer from CT images into four categories: lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), large cell carcinoma (LULC), and normal. Although CNNs have made significant advancements in medical imaging, their limited capacity to capture long-range dependencies has led to the exploration of ViTs, which leverage self-attention mechanisms for a more comprehensive global understanding of images. The study utilized a dataset of 748 lung CT images to train both models with standardized input sizes, assessing their performance through conventional metrics—accuracy, precision, recall, F1 score, specificity, and AUC—as well as cross entropy, a novel metric for evaluating prediction uncertainty. Both models achieved similar accuracy rates (95%), with ViT demonstrating a slight edge over ResNet50 in precision and F1 scores for specific classes. However, ResNet50 exhibited higher recall for LULC, indicating fewer missed cases. Cross entropy analysis showed that the ViT model had lower average uncertainty, particularly in the LUAD, Normal, and LUSC classes, compared to ResNet50. This finding suggests that ViT predictions are generally more reliable, though ResNet50 performed better for LULC. The study underscores that accuracy alone is insufficient for model comparison, as cross entropy offers deeper insights into the reliability and confidence of model predictions. The results highlight the importance of incorporating cross entropy alongside traditional metrics for a more comprehensive evaluation of deep learning models in medical image classification, providing a nuanced understanding of their performance and reliability. While the ViT outperformed the CNN-based ResNet50 in lung cancer classification based on cross-entropy values, the performance differences were minor and may not hold clinical significance. Therefore, it may be premature to consider replacing CNNs with ViTs in this specific application.
基金Project(ZR2014FM036)supported by Shandong Provincial Natural Science Foundation of ChinaProject(ZR2010FZ001)supported by the Key Program of Shandong Provincial Natural Science Foundation of China
文摘In order to increase productivity and reduce energy consumption of steelmaking-continuous casting(SCC) production process, especially with complicated technological routes, the cross entropy(CE) method was adopted to optimize the SCC production scheduling(SCCPS) problem. Based on the CE method, a matrix encoding scheme was proposed and a backward decoding method was used to generate a reasonable schedule. To describe the distribution of the solution space, a probability distribution model was built and used to generate individuals. In addition, the probability updating mechanism of the probability distribution model was proposed which helps to find the optimal individual gradually. Because of the poor stability and premature convergence of the standard cross entropy(SCE) algorithm, the improved cross entropy(ICE) algorithm was proposed with the following improvements: individual generation mechanism combined with heuristic rules, retention mechanism of the optimal individual, local search mechanism and dynamic parameters of the algorithm. Simulation experiments validate that the CE method is effective in solving the SCCPS problem with complicated technological routes and the ICE algorithm proposed has superior performance to the SCE algorithm and the genetic algorithm(GA).
基金Science and Technology Plan of Gansu Province(No.144NKCA040)
文摘In order to improve the global search ability of biogeography-based optimization(BBO)algorithm in multi-threshold image segmentation,a multi-threshold image segmentation based on improved BBO algorithm is proposed.When using BBO algorithm to optimize threshold,firstly,the elitist selection operator is used to retain the optimal set of solutions.Secondly,a migration strategy based on fusion of good solution and pending solution is introduced to reduce premature convergence and invalid migration of traditional migration operations.Thirdly,to reduce the blindness of traditional mutation operations,a mutation operation through binary computation is created.Then,it is applied to the multi-threshold image segmentation of two-dimensional cross entropy.Finally,this method is used to segment the typical image and compared with two-dimensional multi-threshold segmentation based on particle swarm optimization algorithm and the two-dimensional multi-threshold image segmentation based on standard BBO algorithm.The experimental results show that the method has good convergence stability,it can effectively shorten the time of iteration,and the optimization performance is better than the standard BBO algorithm.
基金supported by National Natural Science Foundation of China under Grant No.60872065Open Foundation of State Key Laboratory for Novel Software Technology at Nanjing University under Grant No.KFKT2010B17
文摘The segmentation effect of Tsallis entropy method is superior to that of Shannon entropy method, and the computation speed of two-dimensional Shannon cross entropy method can be further improved by optimization. The existing two-dimensional Tsallis cross entropy method is not the strict two-dimensional extension. Thus two new methods of image thresholding using two-dimensional Tsallis cross entropy based on either Chaotic Particle Swarm Optimization (CPSO) or decomposition are proposed. The former uses CPSO to find the optimal threshold. The recursive algorithm is adopted to avoid the repetitive computation of fitness function in iterative procedure. The computing speed is improved greatly. The latter converts the two-dimensional computation into two one-dimensional spaces, which makes the computational complexity further reduced from O(L2) to O(L). The experimental results show that, compared with the proposed recently two-dimensional Shannon or Tsallis cross entropy method, the two new methods can achieve superior segmentation results and reduce running time greatly.
基金Key Project of Shandong Provincial Natural Science Foundation,China(No.ZR2010FZ001)National High-Tech Research and Development Program of China(863 Program)(No.2007AA04Z157)
文摘A matrix encoding scheme for the steelmaking continuous casting( SCC) production scheduling( SCCPS) problem and the corresponding decoding method are proposed. Based on it,a cross entropy( CE) method is adopted and an improved cross entropy( ICE) algorithm is proposed to solve the SCCPS problem to minimize total power consumption. To describe the distribution of the solution space of the CE method,a probability model is built and used to generate individuals by sampling and a probability updating mechanism is introduced to trace the promising samples. For the ICE algorithm,some samples are generated by the heuristic rules for the shortest makespan due to the relation between the makespan and the total power consumption,which can reduce the search space greatly. The optimal sample in each iteration is retained through a retention mechanism to ensure that the historical optimal sample is not lost so as to improve the efficiency and global convergence. A local search procedure is carried out on a part of better samples so as to improve the local exploitation capability of the ICE algorithm and get a better result. The parameter setting is investigated by the Taguchi method of design-of-experiment. A number of simulation experiments are implemented to validate the effectiveness of the ICE algorithm in solving the SCCPS problem and also the superiority of the ICE algorithm is verified through the comparison with the standard cross entropy( SCE) algorithm.
文摘Emergence of drug resistant bacteria is one of the serious problems in today’s public health. However, the relationship between genomic mutation of bacteria and the phenotypic difference of them is still unclear. In this paper, based on the mutation information in whole genome sequences of 96 MRSA strains, two kinds of phenotypes (pathogenicity and drug resistance) were learnt and predicted by machine learning algorithms. As a result of effective feature selection by cross entropy based sparse logistic regression, these phenotypes could be predicted in sufficiently high accuracy (100% and 97.87%, respectively) with less than 10 features. It means that we could develop a novel rapid test method in the future for checking MRSA phenotypes.
基金the National Natural Science Foundation of China(No.51474100)the Youth Science Fund of Heilongjiang Province in China(No.QC2010023)the Youth Outstanding Ability Program in Heilongjiang University of Science and Technology
文摘To consider multi-objective optimization problem with the number of feed array elements and sidelobe level of large antenna array, multi-objective cross entropy(CE) algorithm is proposed by combining fuzzy c-mean clustering algorithm with traditional cross entropy algorithm, and specific program flow of the algorithm is given.Using the algorithm, large thinned array(200 elements) given sidelobe level(-10,-19 and-30 d B) problem is solved successfully. Compared with the traditional statistical algorithms, the optimization results of the algorithm validate that the number of feed array elements reduces by 51%, 11% and 6% respectively. In addition, compared with the particle swarm optimization(PSO) algorithm, the number of feed array elements from the algorithm is more similar, but the algorithm is more efficient.
基金Supported by the Basic Ability Improvement Project for Young and Middle-aged Teachers in Guangxi Province(2018KY1159).
文摘A health status assessment method based on cross entropy and support vector machine(SVM)is proposed for the new urban rail vehicle traction systems.First,an index system for health assessment of the traction system is established,and combined weights of the index layer are obtained via cross entropy.Then,an SVM assessment model considering actual operating data and each status level of the traction system is established.Finally,the model is simulated in Matlab to obtain assessment results.The results indicate that the proposed method can provide the health status information of the traction system intuitively and complete the health status assessment of the traction system of the new urban rail vehicle effectively,by exploiting the traction system’s layered analysis model.The health status can be assessed accurately and reliably by adopting the cross entropy theory and SVM theory.
文摘In the present study we have formulated a Minimum Cross Fuzzy Entropy Problem (Minx(F)EntP) and proposed sufficient conditions for existence of its solution. Mentioned problem can be formulated as follows. In the set of membership functions satisfying the given moment constraints generated by given moment functions it is required to choose the membership function that is closest to a priori membership function in the sense of cross fuzzy entropy measure. The existence of solution of formulated problem is proved by virtue of concavity property of cross fuzzy entropy measure, the implicit function theorem and Lagrange multipliers method. Moreover, Generalized Cross Fuzzy Entropy Optimization Methods in the form of MinMinx(F)EntM and MaxMinx(F)EntM are suggested on the basis of primary phase of minimizing cross fuzzy entropy measure for fixed moment vector function and on the definition of the special functional with Minx(F)Ent values of cross fuzzy entropy measure. Next phase for obtaining mentioned distributions consists of optimization of defined functional with respect to moment vector functions. Distributions obtained by mentioned methods are defined as (MinMinx(F)Ent)m and (MaxMinx(F)Ent)m distributions.
文摘Value-at-Risk (VaR) estimation via Monte Carlo (MC) simulation is studied here. The variance reduction technique is proposed in order to speed up MC algorithm. The algorithm for estimating the probability of high portfolio losses (more general risk measure) based on the Cross - Entropy importance sampling is developed. This algorithm can easily be applied in any light- or heavy-tailed case without an extra adaptation. Besides, it does not loose in the performance in comparison to other known methods. A numerical study in both cases is performed and the variance reduction rate is compared with other known methods. The problem of VaR estimation using procedures for estimating the probability of high portfolio losses is also discussed.
文摘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.
文摘This study propose a new robust method to rank the performances of multi-assets (portfolios), based purely on their return time series. This method makes no assumption on the distributions. Topsoe distance is symmetrized Kullback-Leibler divergence by average of the probabilities. The square root of Topsoe distance is a metric. We extend this metric from probability density functions to real number series on (0, 1 ]. We call it ST-metric. We show the consistency of ST-metric with mean-variance theory and stochastic dominance method of order one and two. We demonstrate the advantages of ST-metric over mean-variance rule and stochastic dominance method of order one and two.
基金funded by National Natural Science Foundation of China(Grant No. 51805146)the Fundamental Research Funds for the Central Universities (Grant No. B200202221)+1 种基金Jiangsu Key R&D Program (Grant Nos. BE2018004-1, BE2018004)College Students’ Innovative Entrepreneurial Training Plan Program (Grant No. 2020102941513)。
文摘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.
基金supported in part by the Natural Sciences and Engineering Research Council (NSERC) of Canadathe Saskatchewan Power Corporation(SaskPower)。
文摘The increasing penetration of highly intermittent wind generation could seriously jeopardize the operation reliability of power systems and increase the risk of electricity outages. To this end, this paper proposes a novel data-driven method for operation risk assessment of wind-integrated power systems. Firstly, a new approach is presented to model the uncertainty of wind power in lead time. The proposed approach employs k-means clustering and mixture models(MMs) to construct time-dependent probability distributions of wind power.The proposed approach can also capture the complicated statistical features of wind power such as multimodality. Then, a nonsequential Monte Carlo simulation(NSMCS) technique is adopted to evaluate the operation risk indices. To improve the computation performance of NSMCS, a cross-entropy based importance sampling(CE-IS) technique is applied. The CE-IS technique is modified to include the proposed model of wind power.The method is validated on a modified IEEE 24-bus reliability test system(RTS) and a modified IEEE 3-area RTS while employing the historical data of wind generation. The simulation results verify the importance of accurate modeling of shortterm uncertainty of wind power for operation risk assessment.Further case studies have been performed to analyze the impact of transmission systems on operation risk indices. The computational performance of the framework is also examined.