BZPEERSource author record
Explore this author’s publications, collaborators and research topics. This catalog record has not been claimed by its author.
Catalog footprint
Published work
We address pairs (X,Y) of metric spaces with the following property: for every mapping f:X→Y the existence of points x,y∈X with d(f(x),f(y))>d(x,y) implies the existence of x,y∈X for which d(f(x),f(y))<d(x,y). We give sufficient conditions for this property and for its uniform version in terms of finite ε-nets and finite ε-separated subsets.
Let A,X,Y be Banach spaces and A×X→Y, (a,x)↦ax, be a continuous bilinear function, called a *Banach action*. We say that this action *preserves unconditional convergence* if for every bounded sequence (an)n∈ω in A and unconditionally convergent series ∑n∈ωxn in X the series ∑n∈ωanxn is unconditionally convergent. We prove that a Banach action A×X→Y preserves unconditional convergence if and only if for any linear functional y∗∈Y∗ the operator Dy∗:X→A∗, Dy∗(x)(a)=y∗(ax), is absolutely summing. Combining this characterization with the famous Grothendieck theorem on the absolute summability of operators from ℓ1 to ℓ2, we prove that a Banach action A×X→Y preserves unconditional convergence if A is a Hilbert space possessing an orthonormal basis (en)n∈ω such that for every x∈X the series ∑n∈ωenx is weakly absolutely convergent. Applying known results of Garling on the absolute summability of diagonal operators between sequence spaces, we prove that for (finite or infinite) numbers p,q,r∈[1,∞] with r1≤p1+q1, the coordinatewise multplication ℓp×ℓq→ℓr preserves unconditional convergence if and only if one of the following conditions holds: (i) p≤2 and q≤r, (ii) 2<p<q≤r, (iii) 2<p=q<r, (iv) r=∞, (v) 2≤q<p≤r, (vi) q<2<p and p1+q1≥r1+21.
Given a pointed metric space M, we study when there exist n-dimensional linear subspaces of Lip0(M) consisting of strongly norm-attaining Lipschitz functionals, for n∈N. We show that this is always the case for infinite metric spaces, providing a definitive answer to the question. We also study the possible sizes of such infinite-dimensional closed linear subspaces Y, as well as the inverse question, that is, the possible sizes of the metric space M given that such a subspace Y exists. We also show that if the metric space M is σ-precompact, then the aforementioned subspaces Y need to be always separable and isomorphically polyhedral, and we show that for spaces containing [0,1] isometrically, they can be infinite-dimensional.
We prove that for a given Banach space X, the subset of norm attaining Lipschitz functionals in Lip0(X) is weakly dense but not strongly dense. Then we introduce a weaker concept of directional norm attainment and demonstrate that for a uniformly convex X the set of directionally norm attaining Lipschitz functionals is strongly dense in Lip0(X) and, moreover, that an analogue of the Bishop-Phelps-Bollobás theorem is valid.
We prove that an onto isometry between unit spheres of finite-dimensional polyhedral Banach spaces extends to a linear isometry of the corresponding spaces.
We show that for spaces with 1-unconditional bases lushness, the alternative Daugavet property and numerical index~1 are equivalent. In the class of rearrangement invariant (r.i.)\ sequence spaces the only examples of spaces with these properties are c0, ℓ1 and ℓ∞. The only lush r.i.\ separable function space on [0,1] is L1[0,1]; the same space is the only r.i.\ separable function space on [0,1] with the Daugavet property over the reals.
We study those Banach spaces X for which SX does not admit a finite $\eps$-net consisting of elements of SX for any $\eps < 2$. We give characterisations of this class of spaces in terms of ℓ1-type sequences and in terms of the almost Daugavet property. The main result of the paper is: a separable Banach space X is isomorphic to a space from this class if and only if X contains an isomorphic copy of ℓ1.
We consider a general concept of Daugavet property with respect to a norming subspace. This concept covers both the usual Daugavet property and its weak∗ analogue. We introduce and study analogues for narrow operators and rich subspaces in this general setting and apply the results to show that a quotient of L1[0,1] over an ℓ1-subspace can fail the Daugavet property. The latter answers a question posed to us by A. Pelczynski in the negative.
For a compact metric space K the space $\Lip(K)$ has the Daugavet property if and only if the norm of every $f \in \Lip(K)$ is attained locally. If K is a subset of an Lp-space, 1<p<∞, this is equivalent to the convexity of K.
We show that a minor refinement of the Bourgain-Rosenthal construction of a Banach space without the Radon-Nikodym property which contains no bounded δ-trees yields a space with the Daugavet property and the Schur property. Using this example we answer some open questions on the structure of such spaces; in particular we show that the Daugavet property is not inherited by ultraproducts.
We investigate rich subspaces of L1 and deduce an interpolation property of Sidon sets. We also present examples of rich separable subspaces of nonseparable Banach spaces and we study the Daugavet property of tensor products.
We introduce a class of operators on L1 that is stable under taking sums of pointwise unconditionally convergent series, contains all compact operators and does not contain isomorphic embeddings. It follows that any operator from L1 into a space with an unconditional basis belongs to this class.
We characterise narrow and strong Daugavet operators on C(K,E)-spaces; these are in a way the largest sensible classes of operators for which the norm equation ∥Id+T∥=1+∥T∥ is valid. For certain separable range spaces E including all finite-dimensional ones and locally uniformly convex ones we show that an unconditionally pointwise convergent sum of narrow operators on C(K,E) is narrow, which implies for instance the known result that these spaces do not have unconditional FDDs. In a different vein, we construct two narrow operators on C([0,1],ℓ1) whose sum is not narrow.
A Banach space X is said to have the Daugavet property if every operator T:X→X of rank~1 satisfies ∥Id+T∥=1+∥T∥. We show that then every weakly compact operator satisfies this equation as well and that X contains a copy of ℓ1. However, X need not contain a copy of L1. We also study pairs of spaces X⊂Y and operators T:X→Y satisfying ∥J+T∥=1+∥T∥, where J:X→Y is the natural embedding. This leads to the result that a Banach space with the Daugavet property does not embed into a space with an unconditional basis. In another direction, we investigate spaces where the set of operators with ∥Id+T∥=1+∥T∥ is as small as possible and give characterisations in terms of a smoothness condition.
Research graph
Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.
BZPEER is loading the nearby papers, people, topics and institutions for this page.