Bending of Composite Plates under Static and Dynamic Loading

The stationary bending of a two-dimensional elastic system of joined rectangular plates with different mechanical properties and boundary conditions is studied. A technique for solving the corresponding problem is proposed. It is based on approximate approaches in combination with a generalization of the force method. It is established that the static strain state of the system is essentially dependent on the difference between the elastic moduli of the constituent plates. It is also shown that with certain boundary conditions, the dependence of the first resonant frequency on the relative position of plates along the contact line is nonmonotonic__________Translated from Prikladnaya Mekhanika, Vol. 41, No. 1, pp. 77–84, January 2005.     

学习算法的稳定性与泛化(Ⅱ):有界差分稳定框架

根据有界差分条件,提出了学习算法的有界差分稳定框架.依据新框架,研究了机器学习阈值选择算法,再生核Hilbert空间中的正则化学习算法,Ranking学习算法和Bagging算法,证明了对应学习算法的有界差分稳定性.所获结果断言了这些算法均具有有界差分稳定性,从而为这些算法的应用奠定了理论基础.     

统计深度的几何推广

改进统计深度的定义,并将点的深度概念推广到直线与平面的深度,由此得到深度计算的基本定理和深度的一系列性质.最后讨论应用展望.     

Variation of student numerical and figural reasoning approaches by pattern generalization type,strategy use and grade level

This paper explored variation of student numerical and figural reasoning approaches across different pattern generalization types and across grade level. An instrument was designed for this purpose. The instrument was given to a sample of 1232 students from grades 4 to 11 from five schools in Lebanon. Analysis of data showed that the numerical reasoning approach seems to be more dominant than the figural reasoning approach for the near and far pattern generalization types but not for the immediate generalization type. The findings showed that for the recursive strategy, the numerical reasoning approach seems to be more dominant than the figural reasoning approach for each of the three pattern generalization types. However, the figural reasoning approach seems to be more dominant than the numerical reasoning approach for the functional strategy, for each generalization type. The findings also showed that the numerical reasoning was more dominant than the figural reasoning in lower grade levels (grades 4 and 5) for each generalization type. In contrast, the figural reasoning became more dominant than the numerical reasoning in the upper grade levels (grades 10 and 11).     

关于最大公因数的一个定理的推广

本文推广了关于最大公因数的一个定理.     

Information-Theoretic Generalization Bounds for Meta-Learning and Applications

Meta-learning, or “learning to learn”, refers to techniques that infer an inductive bias from data corresponding to multiple related tasks with the goal of improving the sample efficiency for new, previously unobserved, tasks. A key performance measure for meta-learning is the meta-generalization gap, that is, the difference between the average loss measured on the meta-training data and on a new, randomly selected task. This paper presents novel information-theoretic upper bounds on the meta-generalization gap. Two broad classes of meta-learning algorithms are considered that use either separate within-task training and test sets, like model agnostic meta-learning (MAML), or joint within-task training and test sets, like reptile. Extending the existing work for conventional learning, an upper bound on the meta-generalization gap is derived for the former class that depends on the mutual information (MI) between the output of the meta-learning algorithm and its input meta-training data. For the latter, the derived bound includes an additional MI between the output of the per-task learning procedure and corresponding data set to capture within-task uncertainty. Tighter bounds are then developed for the two classes via novel individual task MI (ITMI) bounds. Applications of the derived bounds are finally discussed, including a broad class of noisy iterative algorithms for meta-learning.     

A Generalization on the Difference Between an Integer and Its Inverse Modulo Q

The main purpose of this paper is to use the properties of the Gauss sums, primitive characters and the mean value of Dirichlet L-functions to study the hybrid mean value of the error term E（n, l, c, q） and the hyper-Kloosterman sums K（h,n＋1,q）, the asymptotic property of the mean square value ∑^p c=1 E^2（n, 1, c, p）, and give two interesting mean value formulae.     

中国省域高校科技竞争力演化分析

科技竞争力是国家和区域竞争力的决定性因素。本文将具有时序性特征的全局主成分分析(GPCA)方法引入我国省域高校科技竞争力动态演化研究之中。根据评价方法及评价对象的特点,构建了评价指标体系,并对我国1998-2002年的30个省域高校科技竞争力动态发展演化进行综合分析。结果表明,1998-2002年,我国省域高校科技竞争力发展水平差距缩小,波动明显,竞争激烈。省域经济发展与高校科技竞争力相互关系不确定。     

广义Hojman定理

研究利用Lie对称的生成元τ(t,q,q·)和ξ_s(t,q,q·)来构造广义H ojman守恒 量，并讨论三种特殊情况，研究表明：Hojman守恒量是该广义守恒量的特例，且在Lie对称 的生成元的形式为τ(t,q)和ξ_s(t,q)时，该广义Hojman守恒量可以导出Lu tzky守恒 量，此外，还给出一个排除平凡守恒量的条件.最后，给出两个简单例子，作为所获得结果 的说明. 关键词： 动力学系统 广义Hojman定理 Lie对称 守恒量