与普通网络相比,超网络具有复杂的元组关系(超边),然而现有的大多数网络表示学习方法并不能捕获元组关系。针对上述问题,提出一种超边约束的异质超网络表示学习方法(HRHC)。首先,引入一种结合团扩展和星型扩展的方法,从而将异质超网络...与普通网络相比,超网络具有复杂的元组关系(超边),然而现有的大多数网络表示学习方法并不能捕获元组关系。针对上述问题,提出一种超边约束的异质超网络表示学习方法(HRHC)。首先,引入一种结合团扩展和星型扩展的方法,从而将异质超网络转换为异质网络;其次,引入感知节点语义相关性的元路径游走方法捕获异质节点之间的语义关系;最后,通过超边约束机制捕获节点之间的元组关系,从而获得高质量的节点表示向量。在3个真实世界的超网络数据集上的实验结果表明,对于链接预测任务,所提方法在drug、GPS和MovieLens数据集上都取得了较好的结果;对于超网络重建任务,当超边重建比率大于0.6时,所提方法在drug数据集上的准确性(ACC)优于次优的Hyper2vec(biased 2nd order random walks in Hyper-networks),同时所提方法在GPS数据集上的ACC超过其他基线方法中次优的基于关联图的超边超边约束的异质超网络表示学习方法(HRHC-关联图)15.6个百分点。展开更多
Super-Resolution (SR) technique means to reconstruct High-Resolution (HR) images from a sequence of Low-Resolution (LR) observations,which has been a great focus for compressed video. Based on the theory of Projection...Super-Resolution (SR) technique means to reconstruct High-Resolution (HR) images from a sequence of Low-Resolution (LR) observations,which has been a great focus for compressed video. Based on the theory of Projection Onto Convex Set (POCS),this paper constructs Quantization Constraint Set (QCS) using the quantization information extracted from the video bit stream. By combining the statistical properties of image and the Human Visual System (HVS),a novel Adaptive Quantization Constraint Set (AQCS) is proposed. Simulation results show that AQCS-based SR al-gorithm converges at a fast rate and obtains better performance in both objective and subjective quality,which is applicable for compressed video.展开更多
We calculate one-loop R-parity-violating coupling corrections to theprocesses H~- → τv_τ and H~- → bt. We find that the corrections to the H~- → τv_τ decay modeare generally about 0.1%, and can be negligible. B...We calculate one-loop R-parity-violating coupling corrections to theprocesses H~- → τv_τ and H~- → bt. We find that the corrections to the H~- → τv_τ decay modeare generally about 0.1%, and can be negligible. But the corrections to the H~- → bt decay mode canreach a few percent for the favored parameters.展开更多
The behavior of a thin curved hyperelastic film bonded to a fixed substrate is described by an energy composed of a nonlinearly hyperelastic energy term and a debonding interracial energy term. The author computes the...The behavior of a thin curved hyperelastic film bonded to a fixed substrate is described by an energy composed of a nonlinearly hyperelastic energy term and a debonding interracial energy term. The author computes the Г-limit of this energy under a noninterpenetration constraint that prohibits penetration of the film into the substrate without excluding contact between them.展开更多
A new(γA,σB)-matrix KP hierarchy with two time series γA and σB,which consists of γA-flow,σB-flow and mixed γA and σB-evolution equations of eigenfunctions,is proposed.The reduction and constrained flows of(...A new(γA,σB)-matrix KP hierarchy with two time series γA and σB,which consists of γA-flow,σB-flow and mixed γA and σB-evolution equations of eigenfunctions,is proposed.The reduction and constrained flows of(γA,σB)matrix KP hierarchy are studied.The dressing method is generalized to the(γA,σB)-matrix KP hierarchy and some solutions are presented.展开更多
In this paper,a new modified BFGS method without line searches is proposed.Unlike traditionalBFGS method,this modified BFGS method is proposed based on the so-called fixed steplengthstrategy introduced by Sun and Zhan...In this paper,a new modified BFGS method without line searches is proposed.Unlike traditionalBFGS method,this modified BFGS method is proposed based on the so-called fixed steplengthstrategy introduced by Sun and Zhang.Under some suitable assumptions,the global convergence andthe superlinear convergence of the new algorithm are established,respectively.And some preliminarynumerical experiments,which shows that the new Algorithm is feasible,is also reported.展开更多
This paper proposes an inexact SQP method in association with line search filter technique for solving nonlinear equality constrained optimization. For large-scale applications, it is expensive to get an exact search ...This paper proposes an inexact SQP method in association with line search filter technique for solving nonlinear equality constrained optimization. For large-scale applications, it is expensive to get an exact search direction, and hence the authors use an inexact method that finds an approximate solution satisfying some appropriate conditions. The global convergence of the proposed algorithm is established by using line search filter technique. The second-order correction step is used to overcome the Maratos effect, while the line search filter inexact SQP method has q-superlinear local convergence rate. Finally, the results of numerical experiments indicate that the proposed method is efficient for the given test problems.展开更多
The authors propose a dwindling filter algorithm with Zhou's modified subproblem for nonlinear inequality constrained optimization.The feasibility restoration phase,which is always used in the traditional filter m...The authors propose a dwindling filter algorithm with Zhou's modified subproblem for nonlinear inequality constrained optimization.The feasibility restoration phase,which is always used in the traditional filter method,is not needed.Under mild conditions,global convergence and local superlinear convergence rates are obtained.Numerical results demonstrate that the new algorithm is effective.展开更多
文摘解决约束超多目标优化问题的关键在于约束处理和均衡收敛性与多样性,搜索空间中的约束阻碍种群寻找Pareto前沿面,容易使种群陷入局部最优,而离散的可行域则使种群的多样性较差。提出组合算子型双阶段搜索策略(two-stagesearch strategy with combined operator,TSCO)。TSCO分两阶段处理约束:一阶段算法仅优化目标函数,种群不受约束制约快速向Pareto前沿面方向接近;二阶段通过目标转换将约束违反度视作一个新目标函数以解决原始约束问题。在搜索过程中使用模拟二进制交叉算子和DE/current-to-pbest/1算子构成的组合算子生成收敛性和多样性优秀的个体。为验证策略有效性,结合TSCO策略的AGE-MOEA(TSCOEA)在C_DTLZ、DC_DTLZ和MW测试集上同4种性能优异的约束超多目标进化算法进行对比。实验表明,在大多数问题上,TSCOEA获得的种群收敛性和多样性更好。
文摘与普通网络相比,超网络具有复杂的元组关系(超边),然而现有的大多数网络表示学习方法并不能捕获元组关系。针对上述问题,提出一种超边约束的异质超网络表示学习方法(HRHC)。首先,引入一种结合团扩展和星型扩展的方法,从而将异质超网络转换为异质网络;其次,引入感知节点语义相关性的元路径游走方法捕获异质节点之间的语义关系;最后,通过超边约束机制捕获节点之间的元组关系,从而获得高质量的节点表示向量。在3个真实世界的超网络数据集上的实验结果表明,对于链接预测任务,所提方法在drug、GPS和MovieLens数据集上都取得了较好的结果;对于超网络重建任务,当超边重建比率大于0.6时,所提方法在drug数据集上的准确性(ACC)优于次优的Hyper2vec(biased 2nd order random walks in Hyper-networks),同时所提方法在GPS数据集上的ACC超过其他基线方法中次优的基于关联图的超边超边约束的异质超网络表示学习方法(HRHC-关联图)15.6个百分点。
基金the Natural Science Foundation of Jiangsu Province (No.BK2004151).
文摘Super-Resolution (SR) technique means to reconstruct High-Resolution (HR) images from a sequence of Low-Resolution (LR) observations,which has been a great focus for compressed video. Based on the theory of Projection Onto Convex Set (POCS),this paper constructs Quantization Constraint Set (QCS) using the quantization information extracted from the video bit stream. By combining the statistical properties of image and the Human Visual System (HVS),a novel Adaptive Quantization Constraint Set (AQCS) is proposed. Simulation results show that AQCS-based SR al-gorithm converges at a fast rate and obtains better performance in both objective and subjective quality,which is applicable for compressed video.
文摘We calculate one-loop R-parity-violating coupling corrections to theprocesses H~- → τv_τ and H~- → bt. We find that the corrections to the H~- → τv_τ decay modeare generally about 0.1%, and can be negligible. But the corrections to the H~- → bt decay mode canreach a few percent for the favored parameters.
文摘The behavior of a thin curved hyperelastic film bonded to a fixed substrate is described by an energy composed of a nonlinearly hyperelastic energy term and a debonding interracial energy term. The author computes the Г-limit of this energy under a noninterpenetration constraint that prohibits penetration of the film into the substrate without excluding contact between them.
基金Supported by the National Science Foundation of China under Grant Nos. 10801083,10901090,11171175China Postdoctoral Science Funded Project (20110490408)Chinese Universities Scientific Fund under Grant No. 2011JS041
文摘A new(γA,σB)-matrix KP hierarchy with two time series γA and σB,which consists of γA-flow,σB-flow and mixed γA and σB-evolution equations of eigenfunctions,is proposed.The reduction and constrained flows of(γA,σB)matrix KP hierarchy are studied.The dressing method is generalized to the(γA,σB)-matrix KP hierarchy and some solutions are presented.
基金supported by the Foundation of National Natural Science Foundation of China under Grant No. 10871226the Natural Science Foundation of Shandong Province under Grant No. ZR2009AL006+1 种基金the Development Project Foundation for Science Research of Shandong Education Department under Grant No. J09LA05the Science Project Foundation of Liaocheng University under Grant No. X0810027
文摘In this paper,a new modified BFGS method without line searches is proposed.Unlike traditionalBFGS method,this modified BFGS method is proposed based on the so-called fixed steplengthstrategy introduced by Sun and Zhang.Under some suitable assumptions,the global convergence andthe superlinear convergence of the new algorithm are established,respectively.And some preliminarynumerical experiments,which shows that the new Algorithm is feasible,is also reported.
基金supported by the National Science Foundation Grant under Grant No.10871130the Shanghai Leading Academic Discipline Project under Grant No.T0401
文摘This paper proposes an inexact SQP method in association with line search filter technique for solving nonlinear equality constrained optimization. For large-scale applications, it is expensive to get an exact search direction, and hence the authors use an inexact method that finds an approximate solution satisfying some appropriate conditions. The global convergence of the proposed algorithm is established by using line search filter technique. The second-order correction step is used to overcome the Maratos effect, while the line search filter inexact SQP method has q-superlinear local convergence rate. Finally, the results of numerical experiments indicate that the proposed method is efficient for the given test problems.
基金supported by the National Natural Science Foundation of China(Nos.11201304,11371253)the Innovation Program of Shanghai Municipal Education Commission(No.12YZ174)the Group of Accounting and Governance Disciplines(No.10kq03)
文摘The authors propose a dwindling filter algorithm with Zhou's modified subproblem for nonlinear inequality constrained optimization.The feasibility restoration phase,which is always used in the traditional filter method,is not needed.Under mild conditions,global convergence and local superlinear convergence rates are obtained.Numerical results demonstrate that the new algorithm is effective.
