Let 2≤p【∞ and let (f n) be a martingale. Using exponential bounds of the probabilities of the type P(|f n|】λ‖T(f n)‖ ∞) for some quasi-linear operators acting on martingales, we estimate upper bounds for t...Let 2≤p【∞ and let (f n) be a martingale. Using exponential bounds of the probabilities of the type P(|f n|】λ‖T(f n)‖ ∞) for some quasi-linear operators acting on martingales, we estimate upper bounds for the L p-norms of the maximal functions of martinglaes. Our result is the extension and improvements of the results obtained previously by HITCZENKO and ZENG .展开更多
In surveying data processing,we generally suppose that the observational errors distribute normally.In this case the method of least squares can give the minimum variance unbiased estimation of the parameters.The meth...In surveying data processing,we generally suppose that the observational errors distribute normally.In this case the method of least squares can give the minimum variance unbiased estimation of the parameters.The method of least squares does not have the character of robustness,so the use of it will become unsuitable when a few measurements inheriting gross error mix with others.We can use the robust estimating methods that can avoid the influence of gross errors.With this kind of method there is no need to know the exact distribution of the observations.But it will cause other difficulties such as the hypothesis testing for estimated parameters when the sample size is not so big.For non_normally distributed measurements we can suppose they obey the p _norm distribution law.The p _norm distribution is a distributional class,which includes the most frequently used distributions such as the Laplace,Normal and Rectangular ones.This distribution is symmetric and has a kurtosis between 3 and -6/5 when p is larger than 1.Using p _norm distribution to describe the statistical character of the errors,the only assumption is that the error distribution is a symmetric and unimodal curve.This method possesses the property of a kind of self_adapting.But the density function of the p _norm distribution is so complex that it makes the theoretical analysis more difficult.And the troublesome calculation also makes this method not suitable for practice.The research of this paper indicates that the p _norm distribution can be represented by the linear combination of Laplace distribution and normal distribution or by the linear combination of normal distribution and rectangular distribution approximately.Which kind of representation will be taken is according to whether the parameter p is larger than 1 and less than 2 or p is larger than 2.The approximate distribution have the same first four order moments with the exact one.It means that approximate distribution has the same mathematical expectation,variance,skewness and kurtosis with p _norm distribution.Because every density function used in the approximate formulae has a simple form,using the approximate density function to replace the p _norm ones will simplify the problems of p _norm distributed data processing obviously.展开更多
The cause of the formal difference of p-norm distribution density functions is analyzed, two problems in the deduction of p-norm formulating are improved, and it is proved that two different forms of p-norm distributi...The cause of the formal difference of p-norm distribution density functions is analyzed, two problems in the deduction of p-norm formulating are improved, and it is proved that two different forms of p-norm distribution density functions are equivalent. This work is useful for popularization and application of the p-norm theory to surveying and mapping.展开更多
This paper studies the parameter estimation problems of the nonlinear systems described by the bilinear state space models in the presence of disturbances.A bilinear state observer is designed for deriving identificat...This paper studies the parameter estimation problems of the nonlinear systems described by the bilinear state space models in the presence of disturbances.A bilinear state observer is designed for deriving identification algorithms to estimate the state variables using the input-output data.Based on the bilinear state observer,a novel gradient iterative algorithm is derived for estimating the parameters of the bilinear systems by means of the continuous mixed p-norm cost function.The gain at each iterative step adapts to the data quality so that the algorithm has good robustness to the noise disturbance.Furthermore,to improve the performance of the proposed algorithm,a dynamicmoving window is designed which can update the dynamical data by removing the oldest data and adding the newestmeasurement data.A numerical example of identification of bilinear systems is presented to validate the theoretical analysis.展开更多
For a convex set-valued map between p-normed (0 < p < 1) spaces, we give a criterion for its inverse to be locally Lipschitz of order p. From this we obtain the Robinson-Ursescu Theorem in p-normed spaces and th...For a convex set-valued map between p-normed (0 < p < 1) spaces, we give a criterion for its inverse to be locally Lipschitz of order p. From this we obtain the Robinson-Ursescu Theorem in p-normed spaces and the open mapping and closed graph theorems for closed convex set-valued maps.展开更多
In this paper, using the kernel weight function, we obtain the parameter estimation of p-norm distribution in semi-parametric regression model, which is effective to decide the distribution of random errors. Under the...In this paper, using the kernel weight function, we obtain the parameter estimation of p-norm distribution in semi-parametric regression model, which is effective to decide the distribution of random errors. Under the assumption that the distribution of observations is unimodal and symmetry, this method can give the estimates of the parametric. Finally, two simulated adjustment problem are constructed to explain this method. The new method presented in this paper shows an effective way of solving the problem; the estimated values are nearer to their theoretical ones than those by least squares adjustment approach.展开更多
Negative emotion classification refers to the automatic classification of negative emotion of texts in social networks.Most existing methods are based on deep learning models,facing challenges such as complex structur...Negative emotion classification refers to the automatic classification of negative emotion of texts in social networks.Most existing methods are based on deep learning models,facing challenges such as complex structures and too many hyperparameters.To meet these challenges,in this paper,we propose a method for negative emotion classification utilizing a Robustly Optimized BERT Pretraining Approach(RoBERTa)and p-norm Broad Learning(p-BL).Specifically,there are mainly three contributions in this paper.Firstly,we fine-tune the RoBERTa to adapt it to the task of negative emotion classification.Then,we employ the fine-tuned RoBERTa to extract features of original texts and generate sentence vectors.Secondly,we adopt p-BL to construct a classifier and then predict negative emotions of texts using the classifier.Compared with deep learning models,p-BL has advantages such as a simple structure that is only 3-layer and fewer parameters to be trained.Moreover,it can suppress the adverse effects of more outliers and noise in data by flexibly changing the value of p.Thirdly,we conduct extensive experiments on the public datasets,and the experimental results show that our proposed method outperforms the baseline methods on the tested datasets.展开更多
This paper studies the global exponential p-norm stability of bidirectional associative memory(BAM)neural networks with unbounded time-varying delays.A novel method based on the representation of solutions is put forw...This paper studies the global exponential p-norm stability of bidirectional associative memory(BAM)neural networks with unbounded time-varying delays.A novel method based on the representation of solutions is put forward to deduce a global exponential p-norm stability criterion.This method does not need to set up any Lyapunov-Krasovskii functionals(LKF),which can greatly reduce a large amount of computations and is simpler than the existing methods.In the end,representative numerical examples are given to llustrate the availability of the method.展开更多
By using curvature estimates, we prove that a complete non-compact hypersurface M with constant mean curvature and finite L^n-norm curvature in R^1＋1 must be minimal, so that M is a hyperplane if it is strongly stabl...By using curvature estimates, we prove that a complete non-compact hypersurface M with constant mean curvature and finite L^n-norm curvature in R^1＋1 must be minimal, so that M is a hyperplane if it is strongly stable. This is a generalization of the result on stable complete minimal hypersurfaces of R^n＋1. Moreover, complete strongly stable hypersurfaces with constant mean curvature and finite L^1-norm curvature in R^1＋1 are considered.展开更多
In this paper,a topology optimization method for coordinated stiffness and strength design is proposed under mass constraints,utilizing the Solid Isotropic Material with Penalization approach.Element densities are reg...In this paper,a topology optimization method for coordinated stiffness and strength design is proposed under mass constraints,utilizing the Solid Isotropic Material with Penalization approach.Element densities are regulated through sensitivity filtering tomitigate numerical instabilities associatedwith stress concentrations.Ap-norm aggregation function is employed to globalize local stress constraints,and a normalization technique linearly weights strain energy and stress,transforming the multi-objective problem into a single-objective formulation.The sensitivity of the objective function with respect to design variables is rigorously derived.Three numerical examples are presented,comparing the optimized structures in terms of strain energy,mass,and stress across five different mathematical models with varying combinations of optimization objectives.The results validate the effectiveness and feasibility of the proposed method for achieving a balanced design between structural stiffness and strength.This approach offers a new perspective for future research on stiffness-strength coordinated structural optimization.展开更多
针对核范数正则约束使得矩阵低秩性不足、奇异值分解对大规模数据计算代价大、传统优化算法需人为调试最优参数的问题,提出一种基于Schatten-p范数和近端交替线性最小化算法的深度可学习子空间聚类算法。首先,通过Schatten-p范数作为低...针对核范数正则约束使得矩阵低秩性不足、奇异值分解对大规模数据计算代价大、传统优化算法需人为调试最优参数的问题,提出一种基于Schatten-p范数和近端交替线性最小化算法的深度可学习子空间聚类算法。首先,通过Schatten-p范数作为低秩正则项,使得子空间聚类系数矩阵更好地满足低秩结构;其次,利用Schatten-p范数的矩阵分解格式,避免了SVD计算代价大的不足;最后,针对传统优化算法须人为调整参数的问题,根据激活函数和稀疏正则项的对应关系,建立深度学习网络框架,通过数据自适应学习得到最优参数集。在MNIST手写数字、Amsterdam Library of Object Images和ORL人脸三个数据集上进行聚类的数值实验,结果表明:提出的子空间聚类算法相比于谱聚类、低秩子空间聚类和稀疏子空间聚类算法有更好的聚类性能。展开更多
文摘Let 2≤p【∞ and let (f n) be a martingale. Using exponential bounds of the probabilities of the type P(|f n|】λ‖T(f n)‖ ∞) for some quasi-linear operators acting on martingales, we estimate upper bounds for the L p-norms of the maximal functions of martinglaes. Our result is the extension and improvements of the results obtained previously by HITCZENKO and ZENG .
文摘In surveying data processing,we generally suppose that the observational errors distribute normally.In this case the method of least squares can give the minimum variance unbiased estimation of the parameters.The method of least squares does not have the character of robustness,so the use of it will become unsuitable when a few measurements inheriting gross error mix with others.We can use the robust estimating methods that can avoid the influence of gross errors.With this kind of method there is no need to know the exact distribution of the observations.But it will cause other difficulties such as the hypothesis testing for estimated parameters when the sample size is not so big.For non_normally distributed measurements we can suppose they obey the p _norm distribution law.The p _norm distribution is a distributional class,which includes the most frequently used distributions such as the Laplace,Normal and Rectangular ones.This distribution is symmetric and has a kurtosis between 3 and -6/5 when p is larger than 1.Using p _norm distribution to describe the statistical character of the errors,the only assumption is that the error distribution is a symmetric and unimodal curve.This method possesses the property of a kind of self_adapting.But the density function of the p _norm distribution is so complex that it makes the theoretical analysis more difficult.And the troublesome calculation also makes this method not suitable for practice.The research of this paper indicates that the p _norm distribution can be represented by the linear combination of Laplace distribution and normal distribution or by the linear combination of normal distribution and rectangular distribution approximately.Which kind of representation will be taken is according to whether the parameter p is larger than 1 and less than 2 or p is larger than 2.The approximate distribution have the same first four order moments with the exact one.It means that approximate distribution has the same mathematical expectation,variance,skewness and kurtosis with p _norm distribution.Because every density function used in the approximate formulae has a simple form,using the approximate density function to replace the p _norm ones will simplify the problems of p _norm distributed data processing obviously.
基金Supported by Scientific Research Fund of Hunan Province Education Department (No.03C483) .
文摘The cause of the formal difference of p-norm distribution density functions is analyzed, two problems in the deduction of p-norm formulating are improved, and it is proved that two different forms of p-norm distribution density functions are equivalent. This work is useful for popularization and application of the p-norm theory to surveying and mapping.
基金funded by the National Natural Science Foundation of China(No.61773182)the 111 Project(B12018).
文摘This paper studies the parameter estimation problems of the nonlinear systems described by the bilinear state space models in the presence of disturbances.A bilinear state observer is designed for deriving identification algorithms to estimate the state variables using the input-output data.Based on the bilinear state observer,a novel gradient iterative algorithm is derived for estimating the parameters of the bilinear systems by means of the continuous mixed p-norm cost function.The gain at each iterative step adapts to the data quality so that the algorithm has good robustness to the noise disturbance.Furthermore,to improve the performance of the proposed algorithm,a dynamicmoving window is designed which can update the dynamical data by removing the oldest data and adding the newestmeasurement data.A numerical example of identification of bilinear systems is presented to validate the theoretical analysis.
基金The NSF (Q1107107) of Jiangsu Educational Commission.
文摘For a convex set-valued map between p-normed (0 < p < 1) spaces, we give a criterion for its inverse to be locally Lipschitz of order p. From this we obtain the Robinson-Ursescu Theorem in p-normed spaces and the open mapping and closed graph theorems for closed convex set-valued maps.
文摘In this paper, using the kernel weight function, we obtain the parameter estimation of p-norm distribution in semi-parametric regression model, which is effective to decide the distribution of random errors. Under the assumption that the distribution of observations is unimodal and symmetry, this method can give the estimates of the parametric. Finally, two simulated adjustment problem are constructed to explain this method. The new method presented in this paper shows an effective way of solving the problem; the estimated values are nearer to their theoretical ones than those by least squares adjustment approach.
基金This work was partially supported by the National Natural Science Foundation of China(No.61876205)the Ministry of Education of Humanities and Social Science Project(No.19YJAZH128)+1 种基金the Science and Technology Plan Project of Guangzhou(No.201804010433)the Bidding Project of Laboratory of Language Engineering and Computing(No.LEC2017ZBKT001).
文摘Negative emotion classification refers to the automatic classification of negative emotion of texts in social networks.Most existing methods are based on deep learning models,facing challenges such as complex structures and too many hyperparameters.To meet these challenges,in this paper,we propose a method for negative emotion classification utilizing a Robustly Optimized BERT Pretraining Approach(RoBERTa)and p-norm Broad Learning(p-BL).Specifically,there are mainly three contributions in this paper.Firstly,we fine-tune the RoBERTa to adapt it to the task of negative emotion classification.Then,we employ the fine-tuned RoBERTa to extract features of original texts and generate sentence vectors.Secondly,we adopt p-BL to construct a classifier and then predict negative emotions of texts using the classifier.Compared with deep learning models,p-BL has advantages such as a simple structure that is only 3-layer and fewer parameters to be trained.Moreover,it can suppress the adverse effects of more outliers and noise in data by flexibly changing the value of p.Thirdly,we conduct extensive experiments on the public datasets,and the experimental results show that our proposed method outperforms the baseline methods on the tested datasets.
基金supported in part by the Natural Science Foundation of Heilongjiang Province (No.YQ2021F014)the Fundamental Research Funds for the provincial universities of Heilongjiang Province (No.2020-KYYWF-1040)。
文摘This paper studies the global exponential p-norm stability of bidirectional associative memory(BAM)neural networks with unbounded time-varying delays.A novel method based on the representation of solutions is put forward to deduce a global exponential p-norm stability criterion.This method does not need to set up any Lyapunov-Krasovskii functionals(LKF),which can greatly reduce a large amount of computations and is simpler than the existing methods.In the end,representative numerical examples are given to llustrate the availability of the method.
基金The first author is partially supported by the National Natural Science Foundation of China (No.10271106)The second author is partially supported by the 973-Grant of Mathematics in China and the Huo Y.-D. fund.
文摘By using curvature estimates, we prove that a complete non-compact hypersurface M with constant mean curvature and finite L^n-norm curvature in R^1＋1 must be minimal, so that M is a hyperplane if it is strongly stable. This is a generalization of the result on stable complete minimal hypersurfaces of R^n＋1. Moreover, complete strongly stable hypersurfaces with constant mean curvature and finite L^1-norm curvature in R^1＋1 are considered.
基金funded by National Nature Science Foundation of China(92266203)National Nature Science Foundation of China(52205278)+1 种基金Key Projects of Shijiazhuang Basic Research Program(241791077A)Central Guide Local Science and Technology Development Fund Project of Hebei Province(246Z1022G).
文摘In this paper,a topology optimization method for coordinated stiffness and strength design is proposed under mass constraints,utilizing the Solid Isotropic Material with Penalization approach.Element densities are regulated through sensitivity filtering tomitigate numerical instabilities associatedwith stress concentrations.Ap-norm aggregation function is employed to globalize local stress constraints,and a normalization technique linearly weights strain energy and stress,transforming the multi-objective problem into a single-objective formulation.The sensitivity of the objective function with respect to design variables is rigorously derived.Three numerical examples are presented,comparing the optimized structures in terms of strain energy,mass,and stress across five different mathematical models with varying combinations of optimization objectives.The results validate the effectiveness and feasibility of the proposed method for achieving a balanced design between structural stiffness and strength.This approach offers a new perspective for future research on stiffness-strength coordinated structural optimization.
文摘针对核范数正则约束使得矩阵低秩性不足、奇异值分解对大规模数据计算代价大、传统优化算法需人为调试最优参数的问题,提出一种基于Schatten-p范数和近端交替线性最小化算法的深度可学习子空间聚类算法。首先,通过Schatten-p范数作为低秩正则项,使得子空间聚类系数矩阵更好地满足低秩结构;其次,利用Schatten-p范数的矩阵分解格式,避免了SVD计算代价大的不足;最后,针对传统优化算法须人为调整参数的问题,根据激活函数和稀疏正则项的对应关系,建立深度学习网络框架,通过数据自适应学习得到最优参数集。在MNIST手写数字、Amsterdam Library of Object Images和ORL人脸三个数据集上进行聚类的数值实验,结果表明:提出的子空间聚类算法相比于谱聚类、低秩子空间聚类和稀疏子空间聚类算法有更好的聚类性能。
