Two generalizations of Komlós' theorem with lower closure-type applications.
Se prueba que si f es una aplicación de clase p en un abierto de un cuadrante de un espacio de Banach real, entonces en cada punto del abierto, f admite una extensión de clase p a un entorno global de dicho punto.Se utiliza este resultado para establecer un teorema de extensión de Whitney en un cuadrante de un espacio de Banach y un teorema de la función inversa en variedades con borde anguloso.
A partir de un espacio de Hilbert, E, de dimensión infinita separable y de un elemento λ de L(E,R) - {0} se construye un homeomorfismo h0 de(Eλ+ - Ker λ) U {0}sobre E con las topologías usuales tal que h0(0) = 0 y h0|Eλ+ - Ker λ es un difeomorfismo de clase ∞ de Eλ+ - Ker λ sobre E - {0}, con las estructuras diferenciables de clase ∞ usuales. Mediante h0 se construye una variedad diferenciable de dimensión infinita, separada y no regular.
Let χ(m,n) be the unconditional basis constant of the monomial basis , α ∈ ℕ₀ⁿ with |α| = m, of the Banach space of all m-homogeneous polynomials in n complex variables, endowed with the supremum norm on the n-dimensional unit polydisc ⁿ. We prove that the quotient of and √(n/log n) tends to 1 as n → ∞. This reflects a quite precise dependence of χ(m,n) on the degree m of the polynomials and their number n of variables. Moreover, we give an analogous formula for m-linear forms, a reformulation...
Our aim is to introduce a new notion of unconditionallity, in the context of polynomials in Banach spaces, that looks directly to the polynomial topology defined on the involved spaces. This notion allows us to generalize some well-known relations of duality that appear in the linear context.
The aim of this paper is to show, among other things, that, in separable Banach spaces, the presence of the smoothness with the highest derivative Lipschitzian implies the uniform Gâteaux smoothness of degree 1 up.
Let X be a Banach space, a norming set and (X,B) the topology on X of pointwise convergence on B. We study the following question: given two (non-negative, countably additive and finite) measures μ₁ and μ₂ on Baire(X,w) which coincide on Baire(X,(X,B)), does it follow that μ₁ = μ₂? It turns out that this is not true in general, although the answer is affirmative provided that both μ₁ and μ₂ are convexly τ-additive (e.g. when X has the Pettis Integral Property). For a Banach space Y not containing...