Sobre espacios de Orlicz L(X,μ, E) localmente acotados.
It is shown that every uncountable symmetric basic set in an F-space with a symmetric basis is equivalent to a basic set generated by one vector. We apply this result to investigate the structure of uncountable symmetric basic sets in Orlicz and Lorentz spaces.
It is shown that the order of Gateaux smoothness of bump functions on a wide class of Banach spaces with unconditional basis is not better than that of Fréchet differentiability. It is proved as well that in the separable case this order for Banach lattices satisfying a lower p-estimate for 1≤ p < 2 can be only slightly better.
Several properties of the class of minimal Orlicz function spaces LF are described. In particular, an explicitly defined class of non-trivial minimal functions is shown, which provides concrete examples of Orlicz spaces without complemented copies of F-spaces.
Strictly singular operators on variable exponent (or Nakano) function spaces are characterized in terms of being -strictly singular for the values in the essential range of the exponent function. This extends a result of L. Weiss [On perturbations of Fredholm operators in -spaces, Proc. Amer. Math. Soc. 67 (1977), 287-292] for -spaces.
It is studied when inclusions between rearrangement invariant function spaces on the interval [0,∞) are disjointly strictly singular operators. In particular suitable criteria, in terms of the fundamental function, for the inclusions and to be disjointly strictly singular are shown. Applications to the classes of Lorentz and Marcinkiewicz spaces are given.
If G is the closure of in exp L₂, it is proved that the inclusion between rearrangement invariant spaces E ⊂ F is strictly singular if and only if it is disjointly strictly singular and E ⊊ G. For any Marcinkiewicz space M(φ) ⊂ G such that M(φ) is not an interpolation space between and G it is proved that there exists another Marcinkiewicz space M(ψ) ⊊ M(φ) with the property that the M(ψ) and M(φ) norms are equivalent on the Rademacher subspace. Applications are given and a question of Milman...
In this paper a processing methodology is introduced for the segment or intermediate level in the context of knowledge-based computer vision systems. The proposed methodology demonstrates how using simple Fuzzy Logic concepts it is possible to associate symbolic descriptions to the entities of this level. It provides with the basic mechanisms for performing symbolic computation, evidence combination, uncertainty management and spatial reasoning at the segment level.
Page 1