Subdifferentiability and Inf-Sup Theorems

Subdifferentiability criterions for non necessarily lower semicontinuous convex functions on general locally convex spaces or Fréchet spaces are used to derive inf-sup theorems. The importance of quasicontinuous convex functions is pointed out, and the usual compactness condition relaxed.     

一类Lagrangian对偶问题的零对偶间隙性质及其最优路径的收敛性

针对一般的非线性规划问题,利用某些Lagrange型函数给出了一类Lagrangian对偶问题的一般模型,并证明它与原问题之间存在零对偶间隙．针对具体的一类增广La- grangian对偶问题以及几类由非线性卷积函数构成的Lagrangian对偶问题,详细讨论了零对偶间隙的存在性．进一步,讨论了在最优路径存在的前提下,最优路径的收敛性质．     

Some Results about Duality and Exact Penalization

In this paper, we introduce the concept of the valley at 0 augmenting function and apply it to construct a class of valley at 0 augmented Lagrangian functions. We establish the existence of a path of optimal solutions generated by valley at 0 augmented Lagrangian problems and its convergence toward the optimal set of the original problem and obtain the zero duality gap property between the primal problem and the valley at 0 augmented Lagrangian dual problem. Moreover, we establish the exact penalization representation results in the framework of valley at 0 augmented Lagrangian.     

Zero duality gaps in infinite-dimensional programming

In this paper we study the following infinite-dimensional programming problem: (P) inff ₀(x), subject toxC,f _i(x)0,iI, whereI is an index set with possibly infinite cardinality andC is an infinite-dimensional set. Zero duality gap results are presented under suitable regularity hypotheses for convex-like (nonconvex) and convex infinitely constrained program (P). Various properties of the value function of the convex-like program and its connections to the regularity hypotheses are studied. Relationships between the zero duality gap property, semicontinuity, and -subdifferentiability of the value function are examined. In particular, a characterization for a zero duality gap is given, using the -subdifferential of the value function without convexity.The authors are extremely grateful to the referees for their constructive criticisms and helpful suggestions which have contributed to the final preparation of this paper. This research was partially completed while the first author was a visitor of the Department of Combinatorics and Optimization, University of Waterloo, Waterloo, Ontario, Canada.This work was partially supported by NSERC Grant No. A9161.     

Further Study on Augmented Lagrangian Duality Theory

In this paper, we present a necessary and sufficient condition for a zero duality gap between a primal optimization problem and its generalized augmented Lagrangian dual problems. The condition is mainly expressed in the form of the lower semicontinuity of a perturbation function at the origin. For a constrained optimization problem, a general equivalence is established for zero duality gap properties defined by a general nonlinear Lagrangian dual problem and a generalized augmented Lagrangian dual problem, respectively. For a constrained optimization problem with both equality and inequality constraints, we prove that first-order and second-order necessary optimality conditions of the augmented Lagrangian problems with a convex quadratic augmenting function converge to that of the original constrained program. For a mathematical program with only equality constraints, we show that the second-order necessary conditions of general augmented Lagrangian problems with a convex augmenting function converge to that of the original constrained program.This research is supported by the Research Grants Council of Hong Kong (PolyU B-Q359.)     

非光滑伪线性多目标规划对偶

本文利用Ｄｉｎｉ右上、右下导数给出了非光滑伪线性多目标规划的对偶理论，建立了Ｍｏｎｄ－Ｗｅｉｒ型对仍与Ｗｏｌｆ型对偶；并证明了原问题与对偶问题之间的对偶定理．     

Quality Tolerancing and Conjugate Duality

This paper studies the relationship between product quality as defined by tolerances on the product to tolerances on the parts. All the standard sure-fit and the statistical-fit tolerancing models, which seek to find the least cost tolerances for the parts given the product specification, are found to be convex. Thus conjugate duality provides a unifying framework for studying the relationship between quality and tolerancing and the analysis and solution of such problems. Moreover this convexity is maintained even with the addition of quality loss functions. Quality loss is modeled on the basis of parts or the product. The analysis leads to an analytic solution to a number of models including the quality loss model.     

极小极大分式优化的一个对偶

在函数广义弧连通意义下,建立了极小极大分式优化问题一个对偶模型,并获得了弱对偶和强对偶结果。     

Lower Subdifferentiability and Integration

We consider the question of integration of a multivalued operator T, that is the question of finding a function f such that Tf. If is the Fenchel–Moreau subdifferential, the above problem has been completely solved by Rockafellar, who introduced cyclic monotonicity as a necessary and sufficient condition. In this article we consider the case where f is quasiconvex and is the lower subdifferential ^<. This leads to the introduction of a property that is reminiscent to cyclic monotonicity. We also consider the question of the density of the domains of subdifferential operators.     

Duality gaps in nonconvex stochastic optimization

We consider multistage stochastic optimization models containing nonconvex constraints, e.g., due to logical or integrality requirements. We study three variants of Lagrangian relaxations and of the corresponding decomposition schemes, namely, scenario, nodal and geographical decomposition. Based on convex equivalents for the Lagrangian duals, we compare the duality gaps for these decomposition schemes. The first main result states that scenario decomposition provides a smaller or equal duality gap than nodal decomposition. The second group of results concerns large stochastic optimization models with loosely coupled components. The results provide conditions implying relations between the duality gaps of geographical decomposition and the duality gaps for scenario and nodal decomposition, respectively.Mathematics Subject Classification (1991): 90C15Acknowledgments. This work was supported by the Priority Programme Online Optimization of Large Scale Systems of the Deutsche Forschungsgemeinschaft. The authors wish to thank Andrzej Ruszczyski (Rutgers University) for helpful discussions.     

Duality in the problems of optimal control described by first-order partial differential inclusions

On the basis of the apparatus of locally conjugate mappings, a sufficient condition for optimality is derived for the non-convex problem and duality theorems are proved. A sufficient condition for an extremum is an extremal relation for the direct and dual problem.     

离散型与连续型分布互为对偶的根源

互为对偶的离散型分布与连续型分布,可以看作是由同一个函数——源函数产生的。源函数的正线性组合、乘积和负导数,仍然是源函数。源函数揭示了互为对偶的分布的分布函数之间的相互关系,并能用来求随机变量的数字特征、特征函数、概率母函数、分布的最大值和参数的极大似然估计.     

在pseudo-invexity条件下的一个分式规划问题的对偶性

给出了一个不可微多目标分式变分问题,并利用有效性和真有效性概念,证明了在pseudo-invexity条件下与分式规划问题相关的弱对偶定理、强对偶定理及逆对偶定理.     

Unified Nonlinear Lagrangian Approach to Duality and Optimal Paths

In the context of an inequality constrained optimization problem, we present a unified nonlinear Lagrangian dual scheme and establish necessary and sufficient conditions for the zero duality gap property. From these results, we derive necessary and sufficient conditions for four classes of zero duality gap properties and establish the equivalence among them. Finally, we obtain the convergence of an optimal path for the unified scheme and present a sufficient condition for the finite termination of the optimal path. This research was partially supported by the Research Grants Council of Hong Kong Grant PolyU 5250/03E, the National Natural Science Foundation of China Grants 10471159 and 10571106, NCET, and the Natural Science Foundation of Chongqing     

Metric Distance Function and Profit: Some Duality Results

In this paper, we intend to establish relations between the way efficiency is measured in the literature on efficiency analysis and the notion of distance in topology. To this effect, we are interested particularly in the Hölder norm concept, providing a duality result based upon the profit function. Along this line, we prove that the Luenberger shortage function and the directional distance function of Chambers, Chung, and Färe appear as special cases of some l_p distance (also called Hölder distance), under the assumption that the production set is convex. Under a weaker assumption (convexity of the input correspondence), we derive a duality result based on the cost function, providing several examples in which the functional form of the production set is specified.     

Subdifferentiability and superdifferentiability of distance functions

We obtain necessary and sufficient conditions for the subdifferentiability and superdifferentiability (in the Dem'yanov-Rubinov sense) of the distance in an arbitrary norm from a point to a set for the finitedimensional case. The geometric structure of the subdifferential and the superdifferential is described. Translated fromMatematicheskie Zametki, Vol. 61, No. 4, pp. 530–542, April, 1997. Translated by N. K. Kulman     

Duality of Spectral Curves Arising in Two-Matrix Models

We consider the two-matrix model with the measure given by the exponential of a sum of polynomials in two different variables. We derive a sequence of pairs of dual finite-size systems of ODEs for the corresponding biorthonormal polynomials. We prove an inverse theorem, which shows how to reconstruct such measures from pairs of semi-infinite finite-band matrices, which define the recursion relations and satisfy the string equation. In the limit N, we prove that the obtained dual systems have the same spectral curve.     

Exact Penalty Function and Asymptotic Strong Nonlinear Duality in Integer Programming

In this paper,a logarithmic-exponential penalty function with two parameters for integer program-ming is discussed.We obtain the exact penalty properties and then establish the asymptotic strong nonlinearduality in the corresponding logarithmic-exponential dual formulation by using the obtained exact penaltyproperties.The discussion is based on the logarithmic-exponential nonlinear dual formulation proposed in [6].     

Duality via halfspaces

This note shows that half spaces play a very special role in the development of duality. In addition to the minimum norm duality, the duality in linear programming, and Wolfe's and Johri's formulations in nonlinear programming can all be derived via half spaces by following an identical five step procedure.     

Flag Duality

We introduce a duality on complex flag manifolds that extendsthe usual point-hyperplane duality of complex projective spaces. Thishas consequences for the structure of the linear cycle spaces of flagdomains, especially when those flag domains are not measurable.