On the existence and uniqueness of minimizers for a class of integral functionals

We study the solvability of the minimization problem

Subdifferentiation of integral functionals

Mathematical Programming - We examine how the subdifferentials of nonconvex integral functionals can be deduced from the subdifferentials of the corresponding integrand or at least be estimated...     

Well-posed optimization problems for integral functionals

A necessary and sufficient condition for the correctness of minimum problems defined by integral functionals over pointwise constrained summable functions is that the integrand defines a well-posed minimum problem over the constraint region.This work was supported by the Laboratorio per la Matematica Applicata, Consiglio Nazionale delle Ricerche (CNR).     

On minimum problems for families of functionals

Sommario Per una famiglia { | λ ∈ ⁁} di funzionali reali definiti in uno spazio metricoX e dipendenti da un parametro λ ∈ ⁁, dove ⁁ è uno spazio topologico, si pone il problema di come si comporta, al variare di λ, l'insieme dei punti di minimo di I_λ. Lo studio di tale problema è l'oggetto del presente lavoro e il risultato principale è dato dal Teorema1.

---

---

Entrata in Redazione il 3 marzo 1970.     

Shape derivatives for minima of integral functionals

For \(\Omega \) varying among open bounded sets in \(\mathbb R ^n\) , we consider shape functionals \(J (\Omega )\) defined as the infimum over a Sobolev space of an integral energy of the kind \(\int _\Omega [ f (\nabla u) + g (u) ]\) , under Dirichlet or Neumann conditions on \(\partial \Omega \) . Under fairly weak assumptions on the integrands \(f\) and \(g\) , we prove that, when a given domain \(\Omega \) is deformed into a one-parameter family of domains \(\Omega _\varepsilon \) through an initial velocity field \(V\in W ^ {1, \infty } (\mathbb R ^n, \mathbb R ^n)\) , the corresponding shape derivative of \(J\) at \(\Omega \) in the direction of \(V\) exists. Under some further regularity assumptions, we show that the shape derivative can be represented as a boundary integral depending linearly on the normal component of \(V\) on \(\partial \Omega \) . Our approach to obtain the shape derivative is new, and it is based on the joint use of Convex Analysis and Gamma-convergence techniques. It allows to deduce, as a companion result, optimality conditions in the form of conservation laws.     

Quasi-potentials of the entropy functionals for scalar conservation laws

We investigate the quasi-potential problem for the entropy cost functionals of non-entropic solutions to scalar conservation laws with smooth fluxes. We prove that the quasi-potentials coincide with the integral of a suitable Einstein entropy.     

An existence theory for a minimum energy problem

In this paper we deal with existence theory and develop it for the simple case of the minimum energy problem, as described by Pironneau (1984). We shall treat this problem for the differential inequality by introducing the penalized differential equation and then taking limits of the equations resulting from the penalized approximation.     

An integral for the scalar propagator

Using a convenient contour in the complex plane, an integral representation is obtained for the hypergeometric function, in the case when this function does not reduce to the polynomial case. As an application, a contour integral associated with the massive scalar propagator used in Feynman diagrams is discussed.     

Second-order epi-derivatives of integral functionals

Epi-derivatives have many applications in optimization as approached through nonsmooth analysis. In particular, second-order epi-derivatives can be used to obtain optimality conditions and carry out sensitivity analysis. Therefore the existence of second-order epi-derivatives for various classes of functions is a topic of considerable interest. A broad class of composite functions on ⁿ called fully amenable functions (which include general penalty functions composed withC ² mappings, possibly under a constraint qualification) are now known to be twice epi-differentiable. Integral functionals appear widely in problems in infinite-dimensional optimization, yet to date, only integral functionals defined by convex integrands have been shown to be twice epi-differentiable, provided that the integrands are twice epi-differentiable. Here it is shown that integral functionals are twice epi-differentiable even without convexity, provided only that their defining integrands are twice epi-differentiable and satisfy a uniform lower boundedness condition. In particular, integral functionals defined by fully amenable integrands are twice epi-differentiable under mild conditions on the behavior of the integrands.This work was supported in part by the National Science Foundation under grant DMS-9200303.     

Convex conjugates of integral functionals

We consider a convex integral functional on a functional space V andcompute its greatest extension to the algebraic bidual space V^**, among all convex functions which are lower semicontinuous with respect tothe *-weak topology o(V^** ; V^*).Such computations are usually performed to extend these functionals to sometopological closures. In the present paper, no a priori topological restrictionsare imposed on the extended domain. As a consequence, this extended functionalis a valuable first step for the computation of the exact shape of the minimizersof the conjugate convex integral functional subject to a convex constraint,in full generality: without constraint qualification. These convex integralfunctionals are sometimes called entropies, divergences or energies. Our proofsmainly rely on basic convex duality and duality in Orlicz spaces.     

Approximation of nonlinear integral functionals

Doklady Mathematics -     

Limiting process for integral functionals of a wiener process on a cylinder

We prove that integral functionals, whose integrands are bounded functions of a Wiener process on a cylinder, weakly converge to the processw ₁(τ(t)), τ(t) = β₁ t + (β₂ − β₁)mes {s:w ₂(s)≥0,s<t}, wherew ₁(t andw ₂(t) are independent one-dimensional Wiener processes, β₁ and β₂ are nonrandom values, and β₂≥β₁≥0. Kiev University, Kiev. Translated from Ukrainskii Matematicheskii Zhurnal, Vol. 46, No. 6, pp. 765–768, June, 1994.     

Distribution of integral functionals of a Brownian motion process

In this paper one considers methods which enable one to determine the distribution of certain functionals of a Brownian motion process. Among such functionals we have: the positive continuous additive functional of a Brownian motion, defined by the formula $$A\left( t \right) = \int\limits_{ - \infty }^\infty {\hat t\left( {t, y} \right)dF\left( y \right),} $$ where \(\hat t\left( {t, y} \right)\) is the Brownian local time process while F(y) is a monotonically increasing right continuous function; the functional $$B\left( t \right) = \mathop {\mathop \smallint \limits_{ - \infty } }\nolimits^\infty f\left( {y,\hat t\left( {t, y} \right)} \right)dy,$$ where f(y, x) is a continuous function; and the functional $$C\left( t \right) = \mathop {\mathop \smallint \limits_0 }\nolimits^t f\left( {w\left( s \right),\hat t\left( {sr} \right)} \right)ds$$ As an application of these methods one considers some concrete functionals such that \(\hat t^{ - 1} \left( z \right) = \min \left\{ {s:\hat t\left( {s, o} \right) = z} \right\},\mathop {\mathop \smallint \limits_{ - \infty } }\nolimits^\infty \hat t^2 \left( {t, y} \right)dy,\mathop {\sup }\limits_{y \in R^1 } \hat t\left( {T, y} \right)\) , where T is an exponential random time, independent of \(\hat t\left( {t, y} \right)\) .