On Existence of Nash Equilibria of Games with Constraints on Multistrategies

In this paper, sufficient conditions are given, which are less restrictivethan those required by the Arrow–Debreu–Nash theorem, on theexistence of a Nash equilibrium of an n-player game {₁, . . . , Y_n,f₁, . . . , f_n} in normal form with a nonempty closedconvex constraint C on the set Y=i Y_i of multistrategies. Theith player has to minimize the function fi with respect to the ithvariable. We consider two cases.In the first case, Y is a real Hilbert space and the loss function class isquadratic. In this case, the existence of a Nash equilibrium is guaranteedas a simple consequence of the projection theorem for Hilbert spaces. In thesecond case, Y is a Euclidean space, the loss functions are continuous, andf_i is convex with respect to the ith variable. In this case, the techniqueis quite particular, because the constrained game is approximated with asequence of free games, each with a Nash equilibrium in an appropriatecompact space X. Since X is compact, there exists a subsequence of theseNash equilibrium points which is convergent in the norm. If thelimit point is in C and if the order of convergence is greater than one,then this is a Nash equilibrium of the constrained game.     

Second-order optimality conditions for nondominated solutions of multiobjective programming with C 1,1 data

We examine new second-order necessary conditions and sufficient conditions which characterize nondominated solutions of a generalized constrained multiobjective programming problem. The vector-valued criterion function as well as constraint functions are supposed to be from the class C ^1,1. Second-order optimality conditions for local Pareto solutions are derived as a special case.     

Dynamics on quantum graphs as constrained systems

We study the possibility of regarding the dynamics on a quantum graph as limit, as a small parameter ∈ → O, of a dynamics with a strong confining potential. We define a projection operator along the first eigenfunction of a transversal operator and, under suitable assumptions, we prove that the projection of the solution strongly converges along subsequences to a function that satisfies the Schrödinger equation on each open edge of the graph. Moreover the limit dynamics is unitary. If the limit is independent of the subsequence, one has a limit one-parameter group, generated by one of the self-adjoint extensions of a symmetric operator defined on the open graph (with the vertices deleted). The crucial role of the shape of the confining potential at the vertices is pointed out.     

Energy minimization and molecular dynamics studies of Asn-102 elastase

Summary Four isomeric forms of the Asn-102 PPE (D 102 N mutant according to the emerging protocol, [Knowles, Science, 236 (1987) 1252–1258]) have been investigated using energy minimization (EM) and molecular dynamics (MD) techniques. MD simulation data for 175 ps are reported for each form (in total 700 ps for about 2500 atoms). The His-57 N-protonated forms are calculated to be more stable than the N-protonated ones. The active site region of the most stable form is very similar to that found in the D102N rat trypsin enzyme [Craik et al., Science, 237 (1987) 909–913]. Conformations of the active sites and their hydrogen bond patterns are presented for each of these forms and are compared with the structure of the native enzyme active site. The pH dependent activity of the D102N derivative is discussed.     

Convergence of the Gradient Projection Method for Generalized Convex Minimization

This paper develops convergence theory of the gradient projection method by Calamai and Moré (Math. Programming, vol. 39, 93–116, 1987) which, for minimizing a continuously differentiable optimization problem min{f(x) : x } where is a nonempty closed convex set, generates a sequence x_k+1 = P(x_k – _k f(x_k)) where the stepsize _k > 0 is chosen suitably. It is shown that, when f(x) is a pseudo-convex (quasi-convex) function, this method has strong convergence results: either x_k x^* and x^* is a minimizer (stationary point); or x_k arg min{f(x) : x } = , and f(x_k) inf{f(x) : x }.     

一类凸优化的加速混合下降算法

凸优化问题的混合下降算法利用近似条件的已知信息和随机数扩张预测校正步得到了一组下降方向.而前向加速收缩算法利用高斯赛德尔迭代算法的技术,结合邻近点算法和近似邻近点算法的思想,构造了富有扩张性的下降方向.本文借鉴混合下降算法和前向加速收缩算法的思想,利用已有近似规则信息改善了混合下降算法的下降方向,得到了一类凸优化问题的加速混合下降算法.随后利用Markov不等式、凸函数性质和投影的基本性质等,实现了算法的依概率收敛证明.一系列数值试验表明了加速混合下降算法的有效性和效率性.     

Constrained multiobjective games in locally convex<Emphasis Type="Italic">H</Emphasis>-spaces

A new class of constrained multiobjective games with infinite players in noncompact locally convex H-spaces without linear structure are introduced and studied. By applying a Fan-Glicksberg type fixed point theorem for upper semicontinuous set-valued mappings with closed acyclic values and a maximum theorem, several existence theorems of weighted Nath-equilibria and Pareto equilibria for the constrained multiobjective games are proved in noncompact locally convex H-spaces. These theorems improve, unify and generalize the corresponding results of the multiobjective games in recent literatures. Contributed by Ding Xie-ping Foundation items: the National Natural Science Foundation of China (19871059); the Natural Science Foundation of Education Department of Sichuan Province ([2000]25) Biography: Ding Xie-ping (1938-)     

Waves in Constrained Linear Elastic Materials

This paper deals with the propagation of acceleration waves in constrained linear elastic materials, within the framework of the so-called linearized finite theory of elasticity, as defined by Hoger and Johnson in [12, 13]. In this theory, the constitutive equations are obtained by linearization of the corresponding finite constitutive equations with respect to the displacement gradient and significantly differ from those of the classical linear theory of elasticity. First, following the same procedure used for the constitutive equations, the amplitude condition for a general constraint is obtained. Explicit results for the amplitude condition for incompressible and inextensible materials are also given and compared with those of the classical linear theory of elasticity. In particular, it is shown that for the constraint of incompressibility the classical linear elasticity provides an amplitude condition that, coincidently, is correct, while for the constraint of inextensibility the disagreement is first order in the displacement gradient. Then, the propagation condition for the constraints of incompressibility and inextensibility is studied. For incompressible materials the propagation condition is solved and explicit values for the squares of the speeds of propagation are obtained. For inextensible materials the propagation condition is solved for plane acceleration waves propagating into a homogeneously strained material. For both constraints, it is shown that the squares of the speeds of propagation depend by terms that are first order in the displacement gradient, while in classical linear elasticity they are constant. This revised version was published online in July 2006 with corrections to the Cover Date.     

关于《再论约束最小二乘法》的注记

约束最小二乘法在计算力学和工程问题的理论研究和实际应用中都起到重要的作用。本文借助矩阵广义奇异值分解的方法，研究了一般矩阵的约束最小二乘问题，给出了一般矩阵约束最小二乘问题的算法，推广了文献[1]的结果。