On level sets of marginal functions

The investigation of level sets of marginal functions is motivated by several aspects of standard and generalized semi-infinite programming. The feasible- set M of such a prob­lem is easily seen to be a level set of the marginal function corresponding to the lower level problem. In the present paper we study the local structure of M at feasible bound­ary points in the generic case. A codimension formula shows that there is a wide range of these generic situations, but that the number of active indices is always bounded by the state space dimension. We restrict our attention to two special subcases

In the first case, where the number of active indices is maximalM is shown to be locally diffeomorphic to the non-negative orthant. This situation is well-known from finite and also from standard semi-infinite programming. However, in the second case a generic situation arises which is typical for generalized semi-infinite programming. Here, the active index set is a singleton, and M can exhibit a re-entrant corner or even local non-closedness, depending on whether the Mangasarian-Fromovitz constraint qualifica­tion holds at the active index. If an objective function is minimized over M then in the setting of the second case a local minimizer cannot occur     

The adaptive convexification algorithm for semi-infinite programming with arbitrary index sets

A numerical solution method for semi-infinite optimization problems with arbitrary, not necessarily box-shaped, index sets is presented. Following the ideas of Floudas and Stein (SIAM J Optim 18:1187?C1208, 2007), convex relaxations of the lower level problem are adaptively constructed and then reformulated as mathematical programs with complementarity constraints and solved. Although the index set is arbitrary, this approximation produces feasible iterates for the original problem. The convex relaxations and needed parameters are constructed with ideas of the ??BB method of global optimization and interval methods. It is shown that after finitely many steps an ${\epsilon}$ -stationary point of the original semi-infinite problem is reached. A numerical example illustrates the performance of the proposed method.     

Stability in disjunctive linear optimization I: continuity of the feasible set

The aim of this paper is to study the continuous dependence of the feasible set of a disjunctive semi-infinite linear optimization problem on all involved parameters (matrix and right-hand side). The feasible set of such an optimization problem is the union of (a. possible infinite number of) convex sets, which each is described by a finite or an infinite number of strict and non-strict linear inequalities. We derive necessary and sufficient conditions for the upper- and lower-semi-continuity, and the closedness of the feasible-set-mapping Z Especially, the compactness of the boundary of the feasible set and the closedness of Z are equivalent to the upper-semi-continuity of Zwhile the lower semi-continuity of Z is equivalent to a certain constraint qualification. This constraint qualification is a strengthened kind of Slater condition, rrom tuese investigations, we derive known results in parametric semi-infinite optimization and parametric integer programming.     

Generalized Hadamard well-posedness for infinite vector optimization problems

In this paper, we study the generalized Hadamard well-posedness of infinite vector optimization problems (IVOP). Without the assumption of continuity with respect to the first variable, the upper semicontinuity and closedness of constraint set mappings are established. Under weaker assumptions, sufficient conditions of generalized Hadamard well-posedness for IVOP are obtained under perturbations of both the objective function and the constraint set. We apply our results to the semi-infinite vector optimization problem and the semi-infinite multi-objective optimization problem.     

Instability of closed invariant sets of semidynamical systems: Method of sign-constant Lyapunov functions

In the paper, a reduction principle for the instability property of a closed positively invariant set M for semidynamical systems is proved. The fact that the result is not traditional is stressed by the assumption on the existence of a closed positively invariant set with respect to which the set M has the attraction property. The corresponding instability theorem of the method of sign-constant Lyapunov functions is presented. The assertion thus obtained generalizes the well-known Chetaev and Krasovskii theorems for systems of ordinary differential equations, theorems on the instability with respect to some of the variables, and also the Shimanov and Hale theorems for systems with retarded argument. Illustrating examples are presented.     

Convergence in measure with respect to a matrix-valued measure and some matrix completion problems

For a matrix-valued measure M we introduce a notion of convergence in measure M, which generalizes the notion of convergence in measure with respect to a scalar measure and takes into account the matrix structure of M. Let S be a subset of the set of matrices of given size. It is easy to see that the set of S-valued measurable functions is closed under convergence in measure with respect to a matrix-valued measure if and only if S is a ρ-closed set, i.e. if and only if SP is closed for any orthoprojector P. We discuss the behaviour of ρ-closed sets under operations of linear algebra and the ρ-closedness of particular classes of matrices.     

Generalized semi-infinite optimization: A first order optimality condition and examples

We consider a generalized semi-infinite optimization problem (GSIP) of the form (GSIP) min{f(x)‖xεM}, where M={x∈ℝⁿ|h_i(x)=0i=l,...m, G(x,y)⩾0, y∈Y(x)} and all appearing functions are continuously differentiable. Furthermore, we assume that the setY(x) is compact for allx under consideration and the set-valued mappingY(.) is upper semi-continuous. The difference with a standard semi-infinite problem lies in thex-dependence of the index setY. We prove a first order necessary optimality condition of Fritz John type without assuming a constraint qualification or any kind of reduction approach. Moreover, we discuss some geometrical properties of the feasible setM. This work was partially supported by the “Deutsche Forschungsgemeinschaft” through the Graduiertenkolleg “Mathematische Optimierung” at the University of Trier.     

A new exact exponential penalty function method and nonconvex mathematical programming

A new exact penalty function method, called the l₁ exact exponential penalty function method, is introduced. In this approach, the so-called the exponential penalized optimization problem with the l₁ exact exponential penalty function is associated with the original optimization problem with both inequality and equality constraints. The l₁ exact exponential penalty function method is used to solve nonconvex mathematical programming problems with r-invex functions (with respect to the same function η). The equivalence between sets of optimal solutions of the original mathematical programming problem and of its associated exponential penalized optimization problem is established under suitable r-invexity assumption. Also lower bounds on the penalty parameter are given, for which above these values, this result is true.     

An approximation of feasible sets in semi-infinite optimization

The paper deals with the feasible setM of a semi-infinite optimization problem, i.e.M is a subset of the finite-dimensional Euclidean space and is basically defined by infinitely many inequality constraints. Assuming that the extended Mangasarian-Fromovitz constraint qualification holds at all points fromM, it is shown that the quadratic distance function with respect toM is continuously differentiable on an open neighborhood ofM. If, in addition,M is compact, then the set , which is described by this quadratic distance function, is shown to be an appropriate approximation ofM and the setsM and can be topologically identified via a homeomorphism.     

Jets of closed orbits of Mañé’s generic Hamiltonian flows

We prove a perturbation theorem for the k-jets, k ≥ 2, of the Poincaré map of a closed orbit of the Hamiltonian flow of a Tonelli Hamiltonian H: T *M → ?, on a closed manifold M. As a consequence we obtain Mañé generic properties of Hamiltonian and Lagrangian flows.     

Index sets in the arithmetical Hierarchy

We prove the following results: every recursively enumerable set approximated by finite sets of some set M of recursively enumerable sets with index set in π₂ is an element of M, provided that the finite sets in M are canonically enumerable. If both the finite sets in M and in M̄ are canonically enumerable, then the index set of M is in σ₂∩π₂ if and only if M consists exactly of the sets approximated by finite sets of M and the complement M̄ consists exactly of the sets approximated by finite sets of M̄. Under the same condition M or M̄ has a non-empty subset with recursively enumerable index set, if the index set of M is in σ₂∩π₂.If the finite sets in M are canonically enumerable, then the following three statements are equivalent: (i) the index set of M is in σ₂\π₂, (ii) the index set of M is σ₂-complete, (iii) the index set of M is in σ₂ and some sequence of finite sets in M approximate a set in M̄.Finally, for every n ⩾ 2, an index set in σ_n \ π_n is presented which is not σ_n-complete.     

Infinite distributive laws versus local connectedness and compactness properties

Various local connectedness and compactness properties of topological spaces are characterized by higher degrees of distributivity for their lattices of open (or closed) sets, and conversely. For example, those topological spaces for which not only the lattice of open sets but also that of closed sets is a frame, are described by the existence of web neighborhood bases, where webs are certain specific path-connected sets. Such spaces are called web spaces. The even better linked wide web spaces are characterized by F-distributivity of their topologies, and the worldwide web spaces (or C-spaces) by complete distributivity of their topologies. Similarly, strongly locally connected spaces and locally hypercompact spaces are characterized by suitable infinite distributive laws. The web space concepts are also viewed as natural extensions of spaces that are semilattices with respect to the specialization order and have continuous (unary, binary or infinitary) semilattice operations.     

Constructible Convex Sets

We investigate when closed convex sets can be written as countable intersections of closed half-spaces in Banach spaces. It is reasonable to consider this class to comprise the constructible convex sets since such sets are precisely those that can be defined by a countable number of linear inequalities, hence are accessible to techniques of semi-infinite convex programming. We also explore some model theoretic implications. Applications to set convergence are given as limiting examples.     

Exact Lagrangians in plumbings

Consider a Stein manifold M obtained by plumbing cotangent bundles of manifolds of dimension greater than or equal to 3 at points. We prove that the Fukaya category of closed exact Spin Lagrangians with vanishing Maslov class in M is generated by the compact cores of the plumbing. As applications, we classify exact Lagrangian spheres in A ₂-Milnor fibres of arbitrary dimension, derive constraints on exact Lagrangian fillings of Legendrian unknots in disk cotangent bundles, and prove that the categorical equivalence given by the spherical twist in a homology sphere is typically not realised by any compactly supported symplectomorphism.     

Un théorème de représentation des solutions de viscosité d'une équation d'Hamilton–Jacobi–Bellman ergodique dégénérée sur le tore

We consider an ergodic Hamilton–Jacobi–Bellman equation coming from a stochastic control problem in which there are exactly k points where the dynamics vanishes and the Lagrangian is minimal. Under a stabilizability assumption, we state that the solutions of the ergodic equation are uniquely determined by their value on these k points, and that the set of solutions is sup-norm isometric to a non-empty closed convex set whose dimension is less or equal to k. To cite this article: M. Akian et al., C. R. Acad. Sci. Paris, Ser. I 346 (2008).     

Lower level duality and the global solution of generalized semi-infinite programs

The reformulation of generalized semi-infinite programs (GSIP) to simpler problems is considered. These reformulations are achieved under the assumption that a duality property holds for the lower level program (LLP). Lagrangian duality is used in the general case to establish the relationship between the GSIP and a related semi-infinite program (SIP). Practical aspects of this reformulation, including how to bound the duality multipliers, are also considered. This SIP reformulation result is then combined with recent advances for the global, feasible solution of SIP to develop a global, feasible point method for the solution of GSIP. Reformulations to finite nonlinear programs, and the practical aspects of solving these reformulations globally, are also discussed. When the LLP is a linear program or second-order cone program, specific duality results can be used that lead to stronger results. Numerical examples demonstrate that the global solution of GSIP is computationally practical via the solution of these duality-based reformulations.     

On saddle points in nonconvex semi-infinite programming

In this paper we apply two convexification procedures to the Lagrangian of a nonconvex semi-infinite programming problem. Under the reduction approach it is shown that, locally around a local minimizer, this problem can be transformed equivalently in such a way that the transformed Lagrangian fulfills saddle point optimality conditions, where for the first procedure both the original objective function and constraints (and for the second procedure only the constraints) are substituted by their pth powers with sufficiently large power p. These results allow that local duality theory and corresponding numerical methods (e.g. dual search) can be applied to a broader class of nonconvex problems.     

Poincaré-Hopf inequalities for periodic orbits

The interplay between the dynamics of a nonsingular Morse-Smale flow on a smooth, closed, n-dimensional manifold, M, and the topology of M, was exhibited in Franks (Comment Math Helv 53(2):279?C294, 1978), Smale (Bull Am Math Soc 66:43?C49, 1960), by means of a collection of inequalities, which we refer to as Morse-Smale inequalities. These inequalities relate the number of closed orbits of each index to the Betti numbers of M. These well-known inequalities provide the necessary conditions for a given dynamical data in the form of a specified number of closed orbits of a given index to be realized as a nonsingular Morse-Smale flow on M. In this article we provide two inequalities, hereby referred to as Poincaré-Hopf inequalities for periodic orbits, which imposes constraints on the dynamics of periodic orbits without reference to the Betti numbers of the manifold M. The main theorem establishes the necessity and sufficiency of the Poincaré-Hopf inequalities in order for the Morse-Smale inequalities to hold.     

A class of uniform functions and its relationship with the class of measurable functions

Borel, Lebesgue, and Hausdorff described all uniformly closed families of real-valued functions on a set T whose algebraic properties are just like those of the set of all continuous functions with respect to some open topology on T. These families turn out to be exactly the families of all functions measurable with respect to some σ-additive and multiplicative ensembles on T. The problem of describing all uniformly closed families of bounded functions whose algebraic properties are just like those of the set of all continuous bounded functions remained unsolved. In the paper, a solution of this problem is given with the help of a new class of functions that are uniform with respect to some multiplicative families of finite coverings on T. It is proved that the class of uniform functions differs from the class of measurable functions.     

Viscosity solutions to second order partial differential equations on Riemannian manifolds

We prove comparison, uniqueness and existence results for viscosity solutions to a wide class of fully nonlinear second order partial differential equations F(x,u,du,d²u)=0 defined on a finite-dimensional Riemannian manifold M. Finest results (with hypothesis that require the function F to be degenerate elliptic, that is nonincreasing in the second order derivative variable, and uniformly continuous with respect to the variable x) are obtained under the assumption that M has nonnegative sectional curvature, while, if one additionally requires F to depend on d²u in a uniformly continuous manner, then comparison results are established with no restrictive assumptions on curvature.