In this paper, we consider a discrete-time preemptive priority queue with different service com- pletion probabilities for two classes of customers, one with high-priority and the other with low-priority. This model c...In this paper, we consider a discrete-time preemptive priority queue with different service com- pletion probabilities for two classes of customers, one with high-priority and the other with low-priority. This model corresponds to the classical preemptive priority queueing system with two classes of independent Poisson customers and a single exponential server. Due to the possibility of customers＇ arriving and departing at the same time in a discrete-time queue, the model considered in this paper is more complicated than the continuous- time model. In this model, we focus on the characterization of the exact tail asymptotics for the joint stationary distribution of the queue length of the two types of customers, for the two boundary distributions and for the two marginal distributions, respectively. By using generating functions and the kernel method, we get the exact tail asymptotic properties along the direction of the low-priority queue, as well as along the direction of the high-priority queue.展开更多
Comparing two population proportions using confidence interval could be misleading in many cases, such </span><span style="font-family:Verdana;">as</span><span style="font-family:Ve...Comparing two population proportions using confidence interval could be misleading in many cases, such </span><span style="font-family:Verdana;">as</span><span style="font-family:Verdana;"> the sample size </span><span style="font-family:Verdana;">being</span><span style="font-family:Verdana;"> small and the test </span><span style="font-family:Verdana;">being</span><span style="font-family:Verdana;"> based on normal approximation. In this case, the only </span><span style="font-family:Verdana;">one</span><span style="font-family:Verdana;"> option that we have is to collect a large sample. Unfortunately, the large sample might not be possible. One example is a person suffering from a rare disease. The main purpose of this journal is to derive a closed formula for the exact distribution of the difference between two independent sample proportions, and use it to perform related inferences such as a confidence interval, regardless of the sample sizes and compare with the existing Wald, Agresti-Caffo </span><span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> Score. In this journal, we have derived a closed formula for the exact distribution of the difference between two independent sample proportions. This distribution doesn’t need any </span><span style="font-family:Verdana;">requirements,</span><span style="font-family:Verdana;"> and can be used to perform inferences such </span><span style="font-family:Verdana;">as:</span><span style="font-family:Verdana;"> a hypothesis test for two population proportions, regardless of the nature of the distribution and the sample sizes. We claim </span><span style="font-family:Verdana;">that</span><span style="font-family:Verdana;"> exact distribution has the </span><span style="font-family:Verdana;">least</span><span style="font-family:Verdana;"> confidence width among Wald, Agresti-Caffo </span><span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> Score, so it is suitable for inferences of the difference between the pop展开更多
Test of independence between random vectors X and Y is an essential task in statistical inference.One type of testing methods is based on the minimal spanning tree of variables X and Y.The main idea is to generate the...Test of independence between random vectors X and Y is an essential task in statistical inference.One type of testing methods is based on the minimal spanning tree of variables X and Y.The main idea is to generate the minimal spanning tree for one random vector X,and for each edges in minimal spanning tree,the corresponding rank number can be calculated based on another random vector Y.The resulting test statistics are constructed by these rank numbers.However,the existed statistics are not symmetrical tests about the random vectors X and Y such that the power performance from minimal spanning tree of X is not the same as that from minimal spanning tree of Y.In addition,the conclusion from minimal spanning tree of X might conflict with that from minimal spanning tree of Y.In order to solve these problems,we propose several symmetrical independence tests for X and Y.The exact distributions of test statistics are investigated when the sample size is small.Also,we study the asymptotic properties of the statistics.A permutation method is introduced for getting critical values of the statistics.Compared with the existing methods,our proposed methods are more efficient demonstrated by numerical analysis.展开更多
Aerosol particle deposition in a narrow horizontal channel with gravitational effects is studied in the laminar flow regime. A general explic让 solution for predicting the aerosol transportation in the channel is obta...Aerosol particle deposition in a narrow horizontal channel with gravitational effects is studied in the laminar flow regime. A general explic让 solution for predicting the aerosol transportation in the channel is obtained theoretically. The characteristics of the asymmetric concentratio n profiles and their dependence on gravitational settling are analyzed for different settings of the diffusion and sedimentation parameters. Aerosol deposition on the channel walls, which is influenced by a combination of mechanisms, is also analyzed. The results are compared with those from simplified models obtained in previous studies in which only diffusion or gravity was considered. The analysis reveals that the literature results regarding this aspect may be simplified with the present solution under specific conditions.展开更多
Sticky Brownian motions can be viewed as time-changed semimartingale reflecting Brownian motions,which find applications in many areas including queueing theory and mathematical finance.In this paper,we focus on stati...Sticky Brownian motions can be viewed as time-changed semimartingale reflecting Brownian motions,which find applications in many areas including queueing theory and mathematical finance.In this paper,we focus on stationary distributions for sticky Brownian motions.Main results obtained here include tail asymptotic properties in the marginal distributions and joint distributions.The kernel method,copula concept and extreme value theory are the main tools used in our analysis.展开更多
The ground and low-lying collective states of a rotating system of N=3 bosons harmonically confined in quasi-two-dimension and interacting via repulsive finite-range Gaussian potential is studied in weakly to moderate...The ground and low-lying collective states of a rotating system of N=3 bosons harmonically confined in quasi-two-dimension and interacting via repulsive finite-range Gaussian potential is studied in weakly to moderately interacting regime.The N-body Hamiltonian matrix is diagonalized in subspaces of quantized total angular momenta 0 ≤ L ≤ 4N to obtain the ground and low-lying eigenstates.Our numerical results show that breathing modes with N-body eigenenergy spacing of 2hω⊥,known to exist in strictly 2D system with zero-range(δ-function) interaction potential,may as well exist in quasi-2D system with finite-range Gaussian interaction potential.To gain an insight into the many-body states,the von Neumann entropy is calculated as a measure of quantum correlation and the conditional probability distribution is analyzed for the internal structure of the eigenstates.In the rapidly rotating regime the ground state in angular momentum subspaces L=(q/2)N(N-1) with q=2,4 is found to exhibit the anticorrelation structure suggesting that it may variationally be described by a Bose–Laughlin like state.We further observe that the first breathing mode exhibits features similar to the Bose–Laughlin state in having eigenenergy,von Neumann entropy and internal structure independent of interaction for the three-boson system considered here.On the contrary,for eigenstates lying between the Bose–Laughlin like ground state and the first breathing mode,values of eigenenergy,von Neumann entropy and internal structure are found to vary with interaction.展开更多
Abstract We study the damage probability when M weapons are used against a unitary target. We use the Carleton damage function to model the distribution of damage probability caused by each weapon. The deviation of th...Abstract We study the damage probability when M weapons are used against a unitary target. We use the Carleton damage function to model the distribution of damage probability caused by each weapon. The deviation of the impact point from the aimpoint is attributed to both the dependent error and independent errors. The dependent error is one random variable affecting M weapons the same way while independent errors are associated with individual weapons and are independent of each other. We consider the case where the dependent error is significant, non-negligible relative to independent errors. We first derive an explicit exact solution for the damage probability caused by M weapons for any M. Based on the exact solution, we find the optimal aimpoint distribution of M weapons to maximize the damage probability in several cases where the aimpoint distribution is constrained geometrically with a few free parameters, including uniform distributions around a circle or around an ellipse. Then, we perform unconstrained optimization to obtain the overall optimal aimpoint distribution and the overall maximum damage probability, which is carried out for different values of M, up to 20 weapons. Finally, we derive a phenomenological approximate expression for the damage probability vs. M, the number of weapons, for the parameters studied here.展开更多
无监督域自适应行人重识别(Unsupervised Domain Adaptation for person Re-identification, UDA-ReID)任务致力于将知识从已标记的源域数据转移到目标域。和传统的单源域自适应相比,将多源域的知识迁移到目标域是一项更具挑战性的任务...无监督域自适应行人重识别(Unsupervised Domain Adaptation for person Re-identification, UDA-ReID)任务致力于将知识从已标记的源域数据转移到目标域。和传统的单源域自适应相比,将多源域的知识迁移到目标域是一项更具挑战性的任务。由于领域上的差距,多数据集的简单组合只能产生有限的改进。针对此问题,提出了一种基于精确特征分布匹配和多域信息融合的多源域对比学习(exact feature distribution Matching and multi-domain information Fusion based Multi-domain Contrastive Learning, MFMCL)方法。该方法首先采用具有混合记忆的自步对比学习提取不同域数据的特征,并对提取到的特征进行构图,然后通过两层残差图卷积网络进行多域特征融合。其次,为了增强交叉分布特征、产生更丰富的信息,通过基于排序算法的精确直方图匹配来实现精确特征分布匹配,以获得更多样化的特征增强。实验表明,与目前先进的无监督域自适应行人重识别方法相比,所提出的MFMCL方法在广泛使用的行人重识别数据集Market1501、MSMT17和Duke上都取得了优越的性能。展开更多
Secure group communications are restrained by the number of the group size, number of changes and their distribution, all existing works do not meet the commands of applications with large group size and high dynamic ...Secure group communications are restrained by the number of the group size, number of changes and their distribution, all existing works do not meet the commands of applications with large group size and high dynamic members. In this paper, minimum exact cover problem for group key distribution （GMECP） is presented, and a heuristic solution is testified. Then an algorithm of batch rekeying with renewing cost tending to zero is illustrated, which can process any large number of change requests with best security guaranteed. Efficiency analysis and simulation test show that the achievement can improve the efficiency of any tree-based group key management.展开更多
In this paper, we propose a new generalized p-value for testing homogeneity of scale parameters λi from k independent inverse Gaussian populations. The proposed generalized p-value is proved to have exact frequentist...In this paper, we propose a new generalized p-value for testing homogeneity of scale parameters λi from k independent inverse Gaussian populations. The proposed generalized p-value is proved to have exact frequentist property, and it is also invariant under the group of scale transformation. Simulation results indicate that the proposed test is better than existing approximate χ^2 test.展开更多
基金Supported in part by the National Natural Science Foundation of China under Grant No.11271373 and 11361007the Guangxi Natural Science Foundation under Grant No.2014GXNSFCA118001 and 2012GXNSFBA053010
文摘In this paper, we consider a discrete-time preemptive priority queue with different service com- pletion probabilities for two classes of customers, one with high-priority and the other with low-priority. This model corresponds to the classical preemptive priority queueing system with two classes of independent Poisson customers and a single exponential server. Due to the possibility of customers＇ arriving and departing at the same time in a discrete-time queue, the model considered in this paper is more complicated than the continuous- time model. In this model, we focus on the characterization of the exact tail asymptotics for the joint stationary distribution of the queue length of the two types of customers, for the two boundary distributions and for the two marginal distributions, respectively. By using generating functions and the kernel method, we get the exact tail asymptotic properties along the direction of the low-priority queue, as well as along the direction of the high-priority queue.
文摘Comparing two population proportions using confidence interval could be misleading in many cases, such </span><span style="font-family:Verdana;">as</span><span style="font-family:Verdana;"> the sample size </span><span style="font-family:Verdana;">being</span><span style="font-family:Verdana;"> small and the test </span><span style="font-family:Verdana;">being</span><span style="font-family:Verdana;"> based on normal approximation. In this case, the only </span><span style="font-family:Verdana;">one</span><span style="font-family:Verdana;"> option that we have is to collect a large sample. Unfortunately, the large sample might not be possible. One example is a person suffering from a rare disease. The main purpose of this journal is to derive a closed formula for the exact distribution of the difference between two independent sample proportions, and use it to perform related inferences such as a confidence interval, regardless of the sample sizes and compare with the existing Wald, Agresti-Caffo </span><span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> Score. In this journal, we have derived a closed formula for the exact distribution of the difference between two independent sample proportions. This distribution doesn’t need any </span><span style="font-family:Verdana;">requirements,</span><span style="font-family:Verdana;"> and can be used to perform inferences such </span><span style="font-family:Verdana;">as:</span><span style="font-family:Verdana;"> a hypothesis test for two population proportions, regardless of the nature of the distribution and the sample sizes. We claim </span><span style="font-family:Verdana;">that</span><span style="font-family:Verdana;"> exact distribution has the </span><span style="font-family:Verdana;">least</span><span style="font-family:Verdana;"> confidence width among Wald, Agresti-Caffo </span><span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> Score, so it is suitable for inferences of the difference between the pop
基金Beijing Natural Science Foundation(Grant No.Z200001)National Natural Science Foundation of China(Grant Nos.11871001,11971478 and 11971001)the Fundamental Research Funds for the Central Universities(Grant No.2019NTSS18)。
文摘Test of independence between random vectors X and Y is an essential task in statistical inference.One type of testing methods is based on the minimal spanning tree of variables X and Y.The main idea is to generate the minimal spanning tree for one random vector X,and for each edges in minimal spanning tree,the corresponding rank number can be calculated based on another random vector Y.The resulting test statistics are constructed by these rank numbers.However,the existed statistics are not symmetrical tests about the random vectors X and Y such that the power performance from minimal spanning tree of X is not the same as that from minimal spanning tree of Y.In addition,the conclusion from minimal spanning tree of X might conflict with that from minimal spanning tree of Y.In order to solve these problems,we propose several symmetrical independence tests for X and Y.The exact distributions of test statistics are investigated when the sample size is small.Also,we study the asymptotic properties of the statistics.A permutation method is introduced for getting critical values of the statistics.Compared with the existing methods,our proposed methods are more efficient demonstrated by numerical analysis.
基金the National Natural Science Foundation of China(Grant Nos.51578121,11662017 and 11762019).
文摘Aerosol particle deposition in a narrow horizontal channel with gravitational effects is studied in the laminar flow regime. A general explic让 solution for predicting the aerosol transportation in the channel is obtained theoretically. The characteristics of the asymmetric concentratio n profiles and their dependence on gravitational settling are analyzed for different settings of the diffusion and sedimentation parameters. Aerosol deposition on the channel walls, which is influenced by a combination of mechanisms, is also analyzed. The results are compared with those from simplified models obtained in previous studies in which only diffusion or gravity was considered. The analysis reveals that the literature results regarding this aspect may be simplified with the present solution under specific conditions.
基金supported by the Shandong Provincial Natural Science Foundation of China(Grtant No.ZR2019MA035)the Natural Sciences and Engineering Research Council(NSERC)of Canadasupported by the China Scholarship Council(Grant No.201708370006)。
文摘Sticky Brownian motions can be viewed as time-changed semimartingale reflecting Brownian motions,which find applications in many areas including queueing theory and mathematical finance.In this paper,we focus on stationary distributions for sticky Brownian motions.Main results obtained here include tail asymptotic properties in the marginal distributions and joint distributions.The kernel method,copula concept and extreme value theory are the main tools used in our analysis.
文摘The ground and low-lying collective states of a rotating system of N=3 bosons harmonically confined in quasi-two-dimension and interacting via repulsive finite-range Gaussian potential is studied in weakly to moderately interacting regime.The N-body Hamiltonian matrix is diagonalized in subspaces of quantized total angular momenta 0 ≤ L ≤ 4N to obtain the ground and low-lying eigenstates.Our numerical results show that breathing modes with N-body eigenenergy spacing of 2hω⊥,known to exist in strictly 2D system with zero-range(δ-function) interaction potential,may as well exist in quasi-2D system with finite-range Gaussian interaction potential.To gain an insight into the many-body states,the von Neumann entropy is calculated as a measure of quantum correlation and the conditional probability distribution is analyzed for the internal structure of the eigenstates.In the rapidly rotating regime the ground state in angular momentum subspaces L=(q/2)N(N-1) with q=2,4 is found to exhibit the anticorrelation structure suggesting that it may variationally be described by a Bose–Laughlin like state.We further observe that the first breathing mode exhibits features similar to the Bose–Laughlin state in having eigenenergy,von Neumann entropy and internal structure independent of interaction for the three-boson system considered here.On the contrary,for eigenstates lying between the Bose–Laughlin like ground state and the first breathing mode,values of eigenenergy,von Neumann entropy and internal structure are found to vary with interaction.
文摘Abstract We study the damage probability when M weapons are used against a unitary target. We use the Carleton damage function to model the distribution of damage probability caused by each weapon. The deviation of the impact point from the aimpoint is attributed to both the dependent error and independent errors. The dependent error is one random variable affecting M weapons the same way while independent errors are associated with individual weapons and are independent of each other. We consider the case where the dependent error is significant, non-negligible relative to independent errors. We first derive an explicit exact solution for the damage probability caused by M weapons for any M. Based on the exact solution, we find the optimal aimpoint distribution of M weapons to maximize the damage probability in several cases where the aimpoint distribution is constrained geometrically with a few free parameters, including uniform distributions around a circle or around an ellipse. Then, we perform unconstrained optimization to obtain the overall optimal aimpoint distribution and the overall maximum damage probability, which is carried out for different values of M, up to 20 weapons. Finally, we derive a phenomenological approximate expression for the damage probability vs. M, the number of weapons, for the parameters studied here.
文摘无监督域自适应行人重识别(Unsupervised Domain Adaptation for person Re-identification, UDA-ReID)任务致力于将知识从已标记的源域数据转移到目标域。和传统的单源域自适应相比,将多源域的知识迁移到目标域是一项更具挑战性的任务。由于领域上的差距,多数据集的简单组合只能产生有限的改进。针对此问题,提出了一种基于精确特征分布匹配和多域信息融合的多源域对比学习(exact feature distribution Matching and multi-domain information Fusion based Multi-domain Contrastive Learning, MFMCL)方法。该方法首先采用具有混合记忆的自步对比学习提取不同域数据的特征,并对提取到的特征进行构图,然后通过两层残差图卷积网络进行多域特征融合。其次,为了增强交叉分布特征、产生更丰富的信息,通过基于排序算法的精确直方图匹配来实现精确特征分布匹配,以获得更多样化的特征增强。实验表明,与目前先进的无监督域自适应行人重识别方法相比,所提出的MFMCL方法在广泛使用的行人重识别数据集Market1501、MSMT17和Duke上都取得了优越的性能。
基金Supported by the National Natural Science Foundation of China (60572049)
文摘Secure group communications are restrained by the number of the group size, number of changes and their distribution, all existing works do not meet the commands of applications with large group size and high dynamic members. In this paper, minimum exact cover problem for group key distribution （GMECP） is presented, and a heuristic solution is testified. Then an algorithm of batch rekeying with renewing cost tending to zero is illustrated, which can process any large number of change requests with best security guaranteed. Efficiency analysis and simulation test show that the achievement can improve the efficiency of any tree-based group key management.
基金Supported by the National Natural Science Foundation of China(Grant No.11201478,11471030,11126197 and11471035)
文摘In this paper, we propose a new generalized p-value for testing homogeneity of scale parameters λi from k independent inverse Gaussian populations. The proposed generalized p-value is proved to have exact frequentist property, and it is also invariant under the group of scale transformation. Simulation results indicate that the proposed test is better than existing approximate χ^2 test.
