Convergence for a family of neural network operators in Orlicz spaces

In this paper, we develop the theory for a family of neural network (NN) operators of the Kantorovich type, in the general setting of Orlicz spaces. In particular, a modular convergence theorem is established. In this way, we study the above family of operators in many instances of useful spaces by a unique general approach. The above NN operators provide a constructive approximation process, in which the coefficients, the weights, and the thresholds of the networks needed in order to approximate a given function f , are known. At the end of the paper, several examples of Orlicz spaces, and of sigmoidal activation functions for which the present theory can be applied, are studied in details.     

Lebesgue property for convex risk measures on Orlicz spaces

We present a robust representation theorem for monetary convex risk measures r: X ? \mathbbR{\rho : \mathcal{X} \rightarrow \mathbb{R}} such that

Orlicz capacities and Hausdorff measures on metric spaces

In the setting of doubling metric measure spaces with a 1-Poincaré inequality, we show that sets of Orlicz Φ-capacity zero have generalized Hausdorff h-measure zero provided thatwhere Θ⁻¹ is the inverse of the function Θ(t)=Φ(t)/t, and s is the “upper dimension” of the metric measure space. This condition is a generalization of a well known condition in Rⁿ. For spaces satisfying the weaker q-Poincaré inequality, we obtain a similar but slightly more restrictive condition. Several examples are also provided.     

Convex risk measures on Orlicz spaces: inf-convolution and shortfall

We focus on, throughout this paper, convex risk measures defined on Orlicz spaces. In particular, we investigate basic properties of inf-convolutions defined between a convex risk measure and a convex set, and between two convex risk measures. Moreover, we study shortfall risk measures, which are convex risk measures induced by the shortfall risk. By using results on inf-convolutions, we obtain a robust representation result for shortfall risk measures defined on Orlicz spaces under the assumption that the set of hedging strategies has the sequential compactness in a weak sense. We discuss in addition a construction of an example having the sequential compactness.     

Lattice-universal Orlicz spaces on probability spaces

Lattice-universal Orlicz function spacesL ^F _α,β[0, 1] with prefixed Boyd indices are constructed. Namely, given 0<α<β<∞ arbitrary there exists Orlicz function spacesL ^F _α,β[0, 1] with indices α and β such that every Orlicz function spaceL ^G [0, 1] with indices between α and β is lattice-isomorphic to a sublattice ofL ^F _α,β[0, 1]. The existence of classes of universal Orlicz spacesl ^Fα,β(I) with uncountable symmetric basis and prefixed indices α and β is also proved in the uncountable discrete case. Partially supported by BFM2001-1284.     

Polyhedrality in Orlicz spaces

We present a construction of an Orlicz space admitting a C ^∞-smooth bump which depends locally on finitely many coordinates, and which is not isomorphic to a subspace of any C(K), K scattered. In view of the related results this space is possibly not isomorphic to a polyhedral space. Supported by grants: Institutional Research Plan AV0Z10190503, GAČR 201/04/0090, GAČR 201/07/0394, the research project MSM 0021620839, GAČR 201/05/P582.     

Prediction in Orlicz spaces

In this paper, we investigate the prediction (or best approximation) operator from a uniformly convex real Orlicz space to a subset of -lattice measurable functions. In particular, a counterexample to the monotoncity property, which holds in L_p spaces, is given. Also, a sufficient condition for monotonicity to hold is given. Finally, nested -lattices, as occur in isotonic regression, are considered.This author supported by a grant from the Indiana University-Purdue University at Fort Wayne Research and Instructional Development Support Program     

Pointwise multipliers of Orlicz spaces

We show that the result on multipliers of Orlicz spaces holds in general. Namely, under the assumption that three Young functions Φ₁, Φ₂ and Φ, generating corresponding Orlicz spaces, satisfy the estimate ${\Phi^{-1}(u) \leq C \Phi_1^{-1}(u)\, \Phi_2^{-1}(u)}We show that the result on multipliers of Orlicz spaces holds in general. Namely, under the assumption that three Young functions Φ₁, Φ₂ and Φ, generating corresponding Orlicz spaces, satisfy the estimate F^-1(u) ￡ C F₁^-1(u) F₂^-1(u){\Phi^{-1}(u) \leq C \Phi_1^{-1}(u)\, \Phi_2^{-1}(u)} for all u > 0, we prove that if the pointwise product xy belongs to L ^Φ(μ) for all y ? L^F₁(m){y \in L^{\Phi_1}(\mu)}, then x ? L^F₂(m){x \in L^{\Phi_2}(\mu)}. The result with some restrictions either on Young functions or on the measure μ was proved by Maligranda and Persson (Indag. Math. 51 (1989), 323–338). Our result holds for any collection of three Young functions satisfying the above estimate and for an arbitrary complete σ-finite measure μ.     

On the interpolation constant for Orlicz spaces

In this paper we deal with the interpolation from Lebesgue spaces and , into an Orlicz space , where and for some concave function , with special attention to the interpolation constant . For a bounded linear operator in and , we prove modular inequalities, which allow us to get the estimate for both the Orlicz norm and the Luxemburg norm,

where the interpolation constant depends only on and . We give estimates for , which imply . Moreover, if either or , then . If , then , and, in particular, for the case this gives the classical Orlicz interpolation theorem with the constant .

     

Some fundamental properties for duals of Orlicz spaces

In this paper some basic properties of Orlicz spaces are extended to their dual spaces, and finally, criteria for extreme points in these spaces are given.     

Hardy inequalities in Orlicz spaces

We establish a sharp extension, in the framework of Orlicz spaces, of the (-dimensional) Hardy inequality, involving functions defined on a domain , their gradients and the distance function from the boundary of .

     

Composition operators on Orlicz spaces

In this paper we characterize the composition operators on Orlicz spaces and study some of their properties.Research supported by NBHM DDF No. 40/11/95 (R&D-II)/1429     

Analytic norms in Orlicz spaces

It is shown that an Orlicz sequence space admits an equivalent analytic renorming if and only if it is either isomorphic to or isomorphically polyhedral. As a consequence, we show that there exists a separable Banach space admitting an equivalent -Fréchet norm, but no equivalent analytic norm.