The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention R...The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention Recognition(IR)method for air targets has shortcomings in temporality,interpretability and back-and-forth dependency of intentions.To address these problems,this paper designs a novel air target intention recognition method named STABC-IR,which is based on Bidirectional Gated Recurrent Unit(Bi GRU)and Conditional Random Field(CRF)with Space-Time Attention mechanism(STA).First,the problem of intention recognition of air targets is described and analyzed in detail.Then,a temporal network based on Bi GRU is constructed to achieve the temporal requirement.Subsequently,STA is proposed to focus on the key parts of the features and timing information to meet certain interpretability requirements while strengthening the timing requirements.Finally,an intention transformation network based on CRF is proposed to solve the back-and-forth dependency and transformation problem by jointly modeling the tactical intention of the target at each moment.The experimental results show that the recognition accuracy of the jointly trained STABC-IR model can reach 95.7%,which is higher than other latest intention recognition methods.STABC-IR solves the problem of intention transformation for the first time and considers both temporality and interpretability,which is important for improving the tactical intention recognition capability and has reference value for the construction of command and control auxiliary decision-making system.展开更多
随机森林是机器学习领域中一种常用的分类算法,具有适用范围广且不易过拟合等优点.为了提高随机森林处理多分类问题的能力,提出一种基于空间变换的随机森林算法(space transformation based random forest algorithm,ST-RF).首先,给出...随机森林是机器学习领域中一种常用的分类算法,具有适用范围广且不易过拟合等优点.为了提高随机森林处理多分类问题的能力,提出一种基于空间变换的随机森林算法(space transformation based random forest algorithm,ST-RF).首先,给出一种考虑优先类别的线性判别分析方法(priority class based linear discriminant analysis,PCLDA),利用针对优先类别的投影矩阵对样本进行空间变换,以增强优先类别样本与其他类别样本的区分效果.进而,将PCLDA方法引入随机森林构建过程中,在为每棵决策树随机选择一个优先类别保证随机森林多样性的基础上,利用PCLDA方法创建侧重于不同优先类别的决策树,以提高单棵决策树的分类准确性,从而实现集成模型整体分类性能的有效提升.最后,在10个标准数据集上对ST-RF算法与7种典型随机森林算法进行比较分析,验证所提算法的有效性,并将基于PCLDA的空间变换策略应用到对比算法中,对改进前后的算法性能进行比较分析.实验结果表明:ST-RF算法在处理多分类问题方面具有明显优势,所提出的空间变换策略具有较强的普适性,可以显著提升原算法的分类性能.展开更多
This paper considers the monotonic transformation model with an unspecified transformation function and an unknown error function, and gives its monotone rank estimation with length-biased and rightcensored data. The ...This paper considers the monotonic transformation model with an unspecified transformation function and an unknown error function, and gives its monotone rank estimation with length-biased and rightcensored data. The estimator is shown to be√n-consistent and asymptotically normal. Numerical simulation studies reveal good finite sample performance and the estimator is illustrated with the Oscar data set. The variance can be estimated by a resampling method via perturbing the U-statistics objective function repeatedly.展开更多
In view of the feature of flight flutter test data with atmospheric turbulence excitation, a method which combines wavelet transformation with random decrement technique for identifying flight flutter modal parameters...In view of the feature of flight flutter test data with atmospheric turbulence excitation, a method which combines wavelet transformation with random decrement technique for identifying flight flutter modal parameters is presented. This approach firstly uses random decrement technique to gain free decays corresponding to the acceleration response of the structure to some non-zero initial conditions. Then the continuous Morlet wavelet transformation of the free decays is performed; and the Parseval formula and residue theorem are used to simplify the transformation. The maximal wavelet transformation coefficients in different scales are searched out by means of band-filtering characteristic of Morlet wavelet, and then the modal parameters are identified according to the relationships with maximal modulus and angle of the wavelet transform. In addition, the condition of modal uncoupling is discussed according to variation trend of flight flutter modal parameters in the flight flutter state. The analysis results of simulation and flight flutter test data show that this approach is not only simple, effective and feasible, but also having good noise immunity.展开更多
The requirement of stress analysis and measurement is increasing with the great development of heterogeneous structures and strain engineering in the field of semiconductors.Micro-Raman spectroscopy is an effective me...The requirement of stress analysis and measurement is increasing with the great development of heterogeneous structures and strain engineering in the field of semiconductors.Micro-Raman spectroscopy is an effective method for the measurement of intrinsic stress in semiconductor structures.However,most existing applications of Raman-stress measurement use the classical model established on the (001) crystal plane.A non-negligible error may be introduced when the Raman data are detected on surfaces/cross-sections of different crystal planes.Owing to crystal symmetry,the mechanical,physical and optical parameters of different crystal planes show obvious anisotropy,leading to the Raman-mechanical relationship dissimilarity on the different crystal planes.In this work,a general model of stress measurement on crystalline silicon with an arbitrary crystal plane was presented based on the elastic mechanics,the lattice dynamics and the Raman selection rule.The wavenumberstress factor that is determined by the proposed method is suitable for the measured crystal plane.Detailed examples for some specific crystal planes were provided and the theoretical results were verified by experiments.展开更多
The properties of generalized flip Markov chains on connected regular digraphs are discussed.The 1-Flipper operation on Markov chains for undirected graphs is generalized to that for multi-digraphs.The generalized 1-F...The properties of generalized flip Markov chains on connected regular digraphs are discussed.The 1-Flipper operation on Markov chains for undirected graphs is generalized to that for multi-digraphs.The generalized 1-Flipper operation preserves the regularity and weak connectivity of multi-digraphs.The generalized 1-Flipper operation is proved to be symmetric.Moreover,it is presented that a series of random generalized 1-Flipper operations eventually lead to a uniform probability distribution over all connected d-regular multi-digraphs without loops.展开更多
In this paper,forward expansiveness and entropies of"subsystems"2)of Z^(k)_(+)-actions are investigated.Letαbe a Z^(k)_(+)-action on a compact metric space.For each 1≤j≤k-1,denote G^(j)_(+)={V+:=V∩R^(k)_...In this paper,forward expansiveness and entropies of"subsystems"2)of Z^(k)_(+)-actions are investigated.Letαbe a Z^(k)_(+)-action on a compact metric space.For each 1≤j≤k-1,denote G^(j)_(+)={V+:=V∩R^(k)_(+):V is a j-dimensional subspace of R^(k)}.We consider the forward expansiveness and entropies forαalong V+∈G^(j)_(+).Adapting the technique of"coding",which was introduced by M.Boyle and D.Lind to investigate expansive subdynamics of Z^(k)-actions,to the Z^(k)_(+)cases,we show that the set E^(j)_(+)(α)of forward expansive j-dimensional V_(+)is open in G^(j)_(+).The topological entropy and measure-theoretic entropy of j-dimensional subsystems ofαare both continuous in E^(j)_(+)(α),and moreover,a variational principle relating them is obtained.For a 1-dimensional ray L∈G^(+)_(1),we relate the 1-dimensional subsystem ofαalong L to an i.i.d.random transformation.Applying the techniques of random dynamical systems we investigate the entropy theory of 1-dimensional subsystems.In particular,we propose the notion of preimage entropy(including topological and measure-theoretical versions)via the preimage structure ofαalong L.We show that the preimage entropy coincides with the classical entropy along any L∈E1+(α)for topological and measure-theoretical versions respectively.Meanwhile,a formula relating the measure-theoretical directional preimage entropy and the folding entropy of the generators is obtained.展开更多
In this paper, the concept of Lyapunov exponent is generalized to random transformations that are not necessarily differentiable. For a class of random repellers and of random hyperbolic sets obtained via small pertur...In this paper, the concept of Lyapunov exponent is generalized to random transformations that are not necessarily differentiable. For a class of random repellers and of random hyperbolic sets obtained via small perturbations of deterministic ones respectively, the new exponents are shown to coincide with the classical ones.展开更多
基金supported by the National Natural Science Foundation of China(Nos.62106283 and 72001214)。
文摘The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention Recognition(IR)method for air targets has shortcomings in temporality,interpretability and back-and-forth dependency of intentions.To address these problems,this paper designs a novel air target intention recognition method named STABC-IR,which is based on Bidirectional Gated Recurrent Unit(Bi GRU)and Conditional Random Field(CRF)with Space-Time Attention mechanism(STA).First,the problem of intention recognition of air targets is described and analyzed in detail.Then,a temporal network based on Bi GRU is constructed to achieve the temporal requirement.Subsequently,STA is proposed to focus on the key parts of the features and timing information to meet certain interpretability requirements while strengthening the timing requirements.Finally,an intention transformation network based on CRF is proposed to solve the back-and-forth dependency and transformation problem by jointly modeling the tactical intention of the target at each moment.The experimental results show that the recognition accuracy of the jointly trained STABC-IR model can reach 95.7%,which is higher than other latest intention recognition methods.STABC-IR solves the problem of intention transformation for the first time and considers both temporality and interpretability,which is important for improving the tactical intention recognition capability and has reference value for the construction of command and control auxiliary decision-making system.
文摘随机森林是机器学习领域中一种常用的分类算法,具有适用范围广且不易过拟合等优点.为了提高随机森林处理多分类问题的能力,提出一种基于空间变换的随机森林算法(space transformation based random forest algorithm,ST-RF).首先,给出一种考虑优先类别的线性判别分析方法(priority class based linear discriminant analysis,PCLDA),利用针对优先类别的投影矩阵对样本进行空间变换,以增强优先类别样本与其他类别样本的区分效果.进而,将PCLDA方法引入随机森林构建过程中,在为每棵决策树随机选择一个优先类别保证随机森林多样性的基础上,利用PCLDA方法创建侧重于不同优先类别的决策树,以提高单棵决策树的分类准确性,从而实现集成模型整体分类性能的有效提升.最后,在10个标准数据集上对ST-RF算法与7种典型随机森林算法进行比较分析,验证所提算法的有效性,并将基于PCLDA的空间变换策略应用到对比算法中,对改进前后的算法性能进行比较分析.实验结果表明:ST-RF算法在处理多分类问题方面具有明显优势,所提出的空间变换策略具有较强的普适性,可以显著提升原算法的分类性能.
基金supported by Graduate Innovation Foundation of Shanghai University of Finance and Economics(Grant No.CXJJ2013-451)Cultivation Foundation of Excellent Doctor Degree Dissertation of Shanghai University of Finance and Economics(Grant No.YBPY201504)+4 种基金Program of Educational Department of Fujian Province(Grant Nos.JA14079 and JA12060)Natural Science Foundation of Fujian Province(Grant Nos.2014J01001 and 2012J01028)National Natural Science Foundation of China(Grant No.71271128)the State Key Program of National Natural Science Foundation of China(Grant No.71331006)National Center for Mathematics and Interdisciplinary Sciences,Key Laboratory of Random Complex Structures and Data Science,Chinese Academy of Sciences and Shanghai First-class Discipline A and Innovative Research Team of Shanghai University of Finance and Economics,Program for Changjiang Scholars Innovative Research Team of Ministry of Education(Grant No.IRT13077)
文摘This paper considers the monotonic transformation model with an unspecified transformation function and an unknown error function, and gives its monotone rank estimation with length-biased and rightcensored data. The estimator is shown to be√n-consistent and asymptotically normal. Numerical simulation studies reveal good finite sample performance and the estimator is illustrated with the Oscar data set. The variance can be estimated by a resampling method via perturbing the U-statistics objective function repeatedly.
基金National Natural Science Foundation of China(60134010)
文摘In view of the feature of flight flutter test data with atmospheric turbulence excitation, a method which combines wavelet transformation with random decrement technique for identifying flight flutter modal parameters is presented. This approach firstly uses random decrement technique to gain free decays corresponding to the acceleration response of the structure to some non-zero initial conditions. Then the continuous Morlet wavelet transformation of the free decays is performed; and the Parseval formula and residue theorem are used to simplify the transformation. The maximal wavelet transformation coefficients in different scales are searched out by means of band-filtering characteristic of Morlet wavelet, and then the modal parameters are identified according to the relationships with maximal modulus and angle of the wavelet transform. In addition, the condition of modal uncoupling is discussed according to variation trend of flight flutter modal parameters in the flight flutter state. The analysis results of simulation and flight flutter test data show that this approach is not only simple, effective and feasible, but also having good noise immunity.
基金This work is financially supported by the National Natural Science Foundation of China(Grants 11772223,11772227,and 61727810).
文摘The requirement of stress analysis and measurement is increasing with the great development of heterogeneous structures and strain engineering in the field of semiconductors.Micro-Raman spectroscopy is an effective method for the measurement of intrinsic stress in semiconductor structures.However,most existing applications of Raman-stress measurement use the classical model established on the (001) crystal plane.A non-negligible error may be introduced when the Raman data are detected on surfaces/cross-sections of different crystal planes.Owing to crystal symmetry,the mechanical,physical and optical parameters of different crystal planes show obvious anisotropy,leading to the Raman-mechanical relationship dissimilarity on the different crystal planes.In this work,a general model of stress measurement on crystalline silicon with an arbitrary crystal plane was presented based on the elastic mechanics,the lattice dynamics and the Raman selection rule.The wavenumberstress factor that is determined by the proposed method is suitable for the measured crystal plane.Detailed examples for some specific crystal planes were provided and the theoretical results were verified by experiments.
基金National Natural Science Foundation of China(No.11671258)。
文摘The properties of generalized flip Markov chains on connected regular digraphs are discussed.The 1-Flipper operation on Markov chains for undirected graphs is generalized to that for multi-digraphs.The generalized 1-Flipper operation preserves the regularity and weak connectivity of multi-digraphs.The generalized 1-Flipper operation is proved to be symmetric.Moreover,it is presented that a series of random generalized 1-Flipper operations eventually lead to a uniform probability distribution over all connected d-regular multi-digraphs without loops.
基金Wang and Zhu are supported by NSFC (Grant Nos.11771118,11801336,12171400)Wang is also supported by China Postdoctoral Science Foundation (No.2021M691889)。
文摘In this paper,forward expansiveness and entropies of"subsystems"2)of Z^(k)_(+)-actions are investigated.Letαbe a Z^(k)_(+)-action on a compact metric space.For each 1≤j≤k-1,denote G^(j)_(+)={V+:=V∩R^(k)_(+):V is a j-dimensional subspace of R^(k)}.We consider the forward expansiveness and entropies forαalong V+∈G^(j)_(+).Adapting the technique of"coding",which was introduced by M.Boyle and D.Lind to investigate expansive subdynamics of Z^(k)-actions,to the Z^(k)_(+)cases,we show that the set E^(j)_(+)(α)of forward expansive j-dimensional V_(+)is open in G^(j)_(+).The topological entropy and measure-theoretic entropy of j-dimensional subsystems ofαare both continuous in E^(j)_(+)(α),and moreover,a variational principle relating them is obtained.For a 1-dimensional ray L∈G^(+)_(1),we relate the 1-dimensional subsystem ofαalong L to an i.i.d.random transformation.Applying the techniques of random dynamical systems we investigate the entropy theory of 1-dimensional subsystems.In particular,we propose the notion of preimage entropy(including topological and measure-theoretical versions)via the preimage structure ofαalong L.We show that the preimage entropy coincides with the classical entropy along any L∈E1+(α)for topological and measure-theoretical versions respectively.Meanwhile,a formula relating the measure-theoretical directional preimage entropy and the folding entropy of the generators is obtained.
基金supported by National Natural Science Foundation of China (Grant No. 10701032)Natural Science Foundation of Hebei Province (Grant No. A2008000132)
文摘In this paper, the concept of Lyapunov exponent is generalized to random transformations that are not necessarily differentiable. For a class of random repellers and of random hyperbolic sets obtained via small perturbations of deterministic ones respectively, the new exponents are shown to coincide with the classical ones.
