Searcharxiv⌕ Search

arXiv subjects

Daniel Pellegrino

Publications and source records attributed to Daniel Pellegrino.

At least 55 records · Page 3Linked to original sources

Optimal estimates for summing multilinear operators

We show that given a positive integer $m$, a real number $p\in\left[ 2,\infty\right)$ and $1\leq s<p^{\ast}$ the set of non--multiple $\left( r;s\right)$--summing $m$--linear forms on $\ell_{p}\times\cdots\times \ell_{p}$ contains, except for the null vector, a closed subspace of maximal dimension whenever $r<\frac{2ms}{s+2m-ms}$. This result is optimal since for $r\geq\frac{2ms}{s+2m-ms}$ all $m$--linear forms on $\ell_{p}\times \cdots\times\ell_{p}$ are multiple $\left( r;s\right)$--summing. In particular, among other results, we generalize a result related to cotype (from 2010) due to Botelho \textit{et al.}

math.FA↗

On the size of certain subsets of invariant Banach sequence spaces

The essence of the notion of lineability and spaceability is to find linear structures in somewhat chaotic environments. The existing methods, in general, use \textit{ad hoc} arguments and few general techniques are known. Motivated by the search of general methods, in this paper we formally extend recent results of G.\ Botelho and V.V. Fávaro on invariant sequence spaces to a more general setting. Our main results show that some subsets of invariant sequence spaces contain, up to the null vector, a closed infinite-dimensional subspace.

math.FA↗

Optimal Hardy-Littlewood type inequalities for $m$-linear forms on $\ell_{p}$ spaces with $1\leq p\leq m$

The Hardy-Littlewood inequalities for $m$-linear forms on $\ell_{p}$ spaces are stated for $p>m$. In this paper, among other results, we investigate similar results for $1\leq p\leq m.$ Let $\mathbb{K}$ be $% \mathbb{R}$ or $\mathbb{C}$ and $m\geq 2$ be a positive integer. Our main results are the following sharp inequalities: (i) If $\left(r,p\right) \in \left(\lbrack 1,2]\times \lbrack 2,2m)\right) \cup \left(\lbrack 1,\infty)\times \lbrack 2m,\infty \right)) $, then there is a constant $D_{m,r,p}^{\mathbb{K}}>0$ (not depending on $% n $) such that \begin{equation*} \textstyle\left(\sum\limits_{j_{1},...,j_{m}=1}^{n}\left\vert T(e_{j_{1}},...,e_{j_{m}})\right\vert ^{r}\right) ^{\frac{1}{r}}\leq D_{m,r,p}^{\mathbb{K}}n^{\max \left\{ \frac{2mr+2mp-mpr-pr}{2pr},0\right\} }\left\Vert T\right\Vert \end{equation*} for all $m$--linear forms $T:\ell_{p}^{n}\times \cdots \times \ell_{p}^{n}\rightarrow \mathbb{K}$ and all positive integers $n$. (ii) If $\left(r,p\right) \in \lbrack 2,\infty)\times (m,2m]$, then \begin{equation*} \textstyle\left(\sum\limits_{j_{1},...,j_{m}=1}^{n}\left\vert T(e_{j_{1}},...,e_{j_{m}})\right\vert ^{r}\right) ^{\frac{1}{r}}\leq \left(\sqrt{2}\right) ^{m-1}n^{\max \left\{ \frac{p+mr-rp}{pr},0\right\} }\left\Vert T\right\Vert \end{equation*} for all $m$--linear forms $T:\ell_{p}^{n}\times \cdots \times \ell_{p}^{n}\rightarrow \mathbb{K}$ and all positive integers $n.$ Moreover the exponents $\max \{ (2mr+2mp-mpr-pr)/2pr,0 \} $ in (i) and $\max \{(p+mr-rp)/pr,0 \} $ in (ii) are optimal.

math.FA↗

Improving integrability via absolute summability: a general version of Diestel's Theorem

A classical result by J. Diestel establishes that the composition of a summing operator with a (strongly measurable) Pettis integrable function gives a Bochner integrable function. In this paper we show that a much more general result is possible regarding the improvement of the integrability of vector valued functions by the summability of the operator. After proving a general result, we center our attention in the particular case given by the $(p,σ)$-absolutely continuous operators, that allows to prove a lot of special results on integration improvement for selected cases of classical Banach spaces ---including $C(K)$, $L^p$ and Hilbert spaces--- and operators ---$p$-summing, $(q,p)$-summing and $p$-approximable operators---.

math.FA↗

Lineability and uniformly dominated sets of summing nonlinear operators

In this note we prove an abstract version of a result from 2002 due to Delgado and Piñero on absolutely summing operators. Several applications are presented; some of them in the multilinear framework and some in a completely nonlinear setting. In a final section we investigate the size of the set of non uniformly dominated sets of linear operators under the point of view of lineability.

math.FA↗

Lower bounds for the complex polynomial Hardy--Littlewood inequality

The Hardy--Littlewood inequality for complex homogeneous polynomials asserts that given positive integers $m\geq2$ and $n\geq1$, if $P$ is a complex homogeneous polynomial of degree $m$ on $\ell_{p}^{n}$ with $2m\leq p\leq\infty$ given by $P(x_{1},\ldots,x_{n})=\sum_{|α|=m}a_{α}\mathbf{{x}^α}$, then there exists a constant $C_{\mathbb{C},m,p}^{\mathrm{pol}}\geq1$ (which is does not depend on $n$) such that \[ \left( {\sum\limits_{\left\vert α\right\vert =m}}\left\vert a_{α}\right\vert ^{\frac{2mp}{mp+p-2m}}\right) ^{\frac{mp+p-2m}{2mp}}\leq C_{\mathbb{C},m,p}^{\mathrm{pol}}\left\Vert P\right\Vert , \] with $\Vert P\Vert:=\sup_{z\in B_{\ell_{p}^{n}}}|P(z)|$. In this short note, among other results, we provide nontrivial lower bounds for the constants $C_{\mathbb{C},m,p}^{\mathrm{pol}}$. For instance we prove that, for $m\geq2$ and $2m\leq p<\infty$, \[ C_{\mathbb{C},m,p}^{\mathrm{pol}}\geq2^{\frac{m}{p}}% \] for $m$ even, and \[ C_{\mathbb{C},m,p}^{\mathrm{pol}}\geq2^{\frac{m-1}{p}}% \] for $m$ odd. Estimates for the case $p=\infty$ (this is the particular case of the complex polynomial Bohnenblust--Hille inequality) were recently obtained by D. Nuñez-Alarcón in 2013.

math.FA↗

Lower bounds for the constants of the Hardy-Littlewood inequalities

Given an integer $m\geq2$, the Hardy--Littlewood inequality (for real scalars) says that for all $2m\leq p\leq\infty$, there exists a constant $C_{m,p}% ^{\mathbb{R}}\geq1$ such that, for all continuous $m$--linear forms $A:\ell_{p}^{N}\times\cdots\times\ell_{p}^{N}\rightarrow\mathbb{R}$ and all positive integers $N$, \[ \left( \sum_{j_{1},...,j_{m}=1}^{N}\left\vert A(e_{j_{1}},...,e_{j_{m}% })\right\vert ^{\frac{2mp}{mp+p-2m}}\right) ^{\frac{mp+p-2m}{2mp}}\leq C_{m,p}^{\mathbb{R}}\left\Vert A\right\Vert . \] The limiting case $p=\infty$ is the well-known Bohnenblust--Hille inequality; the behavior of the constants $C_{m,p}^{\mathbb{R}}$ is an open problem. In this note we provide nontrivial lower bounds for these constants.

math.FA↗

On the constants of the Bohnenblust-Hille inequality and Hardy--Littlewood inequalities

In this paper, among other results, we improve the best known estimates for the constants of the generalized Bohnenblust-Hille inequality. These enhancements are then used to improve the best known constants of the Hardy--Littlewood inequality; this inequality asserts that for a positive integer $m\geq2$ with $2m\leq p\leq\infty$ and $\mathbb{K}=\mathbb{R}$ or $\mathbb{C}$ there exists a constant $C_{m,p}^{\mathbb{K}}\geq1$ such that, for all continuous $m$--linear forms $T:\ell_{p}^{n}\times\cdots\times\ell_{p}^{n}\rightarrow\mathbb{K}$, and all positive integers $n$,% \[ \left( \sum_{j_{1},...,j_{m}=1}^{n}\left\vert T(e_{j_{1}},...,e_{j_{m}% })\right\vert ^{\frac{2mp}{mp+p-2m}}\right) ^{\frac{mp+p-2m}{2mp}}\leq C_{m,p}^{\mathbb{K}}\left\Vert T\right\Vert , \] and the exponent $\frac{2mp}{mp+p-2m}$ is sharp. In particular, we show that for $p > 2m^{3}-4m^{2}+2m$ the optimal constants satisfying the above inequality are dominated by the best known estimates for the constants of the $m$-linear Bohnenblust--Hille inequality. More precisely if $γ$ denotes the Euler--Mascheroni constant, considering the case of complex scalars as an illustration, we show that% \[ C_{m,p}^{\mathbb{C}}\leq\prod\limits_{j=2}^{m}Γ\left( 2-\frac{1}% {j}\right) ^{\frac{j}{2-2j}}<m^{\frac{1-γ}{2}}, \] which is somewhat surprising since this new formula has no dependence on $p$ (the former estimate depends on $p$ but, paradoxally, is worse than this new one). This suggests the following open problems: 1) Are the optimal constants of the Hardy--Littlewood inequality and Bohnenblust--Hille inequalities the same? 2) Are the optimal constants of the Hardy--Littlewood inequality independent of $p$ (at least for large $p$)?

math.FA↗

Optimal Hardy-Littlewood type inequalities for polynomials and multilinear operators

In this paper we obtain quite general and definitive forms for Hardy-Littlewood type inequalities. Moreover, when restricted to the original particular cases, our approach provides much simpler and straightforward proofs and we are able to show that in most cases the exponents involved are optimal. The technique we used is a combination of probabilistic tools and of an interpolative approach; this former technique is also employed in this paper to improve the constants for vector-valued Bohnenblust--Hille type inequalities.

math.FA↗

On the upper bounds for the constants of the Hardy-Littlewood inequality

The best known upper estimates for the constants of the Hardy--Littlewood inequality for $m$-linear forms on $\ell_{p}$ spaces are of the form $\left(\sqrt{2}\right) ^{m-1}.$ We present better estimates which depend on $p$ and $m$. An interesting consequence is that if $p\geq m^{2}$ then the constants have a subpolynomial growth as $m$ tends to infinity.

math.FA↗

Coincidence results for summing multilinear mappings

In this paper we prove coincidence results concerning spaces of absolutely summing multilinear mappings between Banach spaces. The nature of these results arises from two distinct approaches: the coincidence of two \textit{a priori} different classes of summing multilinear mappings and the summability of all multilinear mappings defined on products of Banach spaces. Optimal generalizations of known results are obtained. We also introduce and explore new techniques in the field, for example a technique to extend coincidence results for linear, bilinear and even trilinear mappings to general multilinear ones.

math.FA↗

Grothendieck's theorem for absolutely summing multilinear operators is optimal

Grothendieck's theorem asserts that every continuous linear operator from $\ell_{1}$ to $\ell_{2}$ is absolutely $\left( 1;1\right) $-summing. In this note we prove that the optimal constant $g_{m}$ so that every continuous $m$-linear operator from $\ell_{1}\times\cdots\times\ell_{1}$ to $\ell_{2}$ is absolutely $\left( g_{m};1\right) $-summing is $\frac{2}{m+1}$. We also show that if $g_{m}<\frac{2}{m+1}$ there is $\mathfrak{c}$ dimensional linear space composed by continuous non absolutely $\left( g_{m};1\right) $-summing $m$-linear operators from $\ell_{1}\times\cdots\times\ell_{1}$ to $\ell_{2}.$ In particular, our result solves (in the positive) a conjecture posed by A.T. Bernardino in 2011.

math.FA↗

Weak compactness and strongly summing multilinear operators

Every absolutely summing linear operator is weakly compact. However, for strongly summing multilinear operators and polynomials - one of the most natural extensions of the linear case to the non linear framework - weak compactness does not hold in general. We show that a subclass of the class of strongly summing multilinear operators/polynomials, sharing its main properties such as Grothendieck's Theorem, Pietsch Domination Theorem and Dvoretzky-Rogers Theorem, has even better properties like weak compactness and a natural factorization theorem.

math.FA↗

An interpolation technique towards the subpolynomial constants in the multilinear Bohnenblust-Hille inequality

We show that a recent interpolative new proof of the Bohnenblust--Hille inequality, when suitably handled, recovers its best known constants. This seems to be unexpectedly surprising since the known interpolative approaches only provide constants having exponential growth. This preprint is no longer an independent submission, it is now contained in the preprint arXiv 1310.2834.

math.FA↗

Lineability, spaceability, and additivity cardinals for Darboux-like functions

We introduce the concept of {\em maximal lineability cardinal number}, $\mL(M)$, of a subset $M$ of a topological vector space and study its relation to the cardinal numbers known as: additivity $A(M)$, homogeneous lineability $\HL(M)$, and lineability $\LL(M)$ of $M$. In particular, we will describe, in terms of $\LL$, the lineability and spaceability of the families of the following Darboux-like functions on $\real^n$, $n\ge 1$: extendable, Jones, and almost continuous functions.

math.FA↗

A note on the hypercontractivity of the polynomial Bohnenblust--Hille inequality

For $\mathbb{K}=\mathbb{R}$ or $\mathbb{C}$ and $m$ a positive integer, we remark that there is a constant $C$ so that, for all $r\in\lbrack1,\frac {2m}{m+1}],$ the supremum of the ratio between the $\ell_{r}$ norm of the coefficients of any nonzero $m$-homogeneous polynomial $P:\ell_{\infty}% ^{n}(\mathbb{K}) \rightarrow\mathbb{K}$ and its supremum norm is dominated by $C^{m}\cdot n^{(\frac{m}{r}-\frac{m+1}{2})}$ and, moreover, we prove that the exponent $\frac{m}{r}-\frac{m+1}{2}$ is optimal.

math.FA↗

On Pietsch measures for summing operators and dominated polynomials

We relate the injectivity of the canonical map from $C(B_{E'})$ to $L_p(μ)$, where $μ$ is a regular Borel probability measure on the closed unit ball $B_{E'}$ of the dual $E'$ of a Banach space $E$ endowed with the weak* topology, to the existence of injective $p$-summing linear operators/$p$-dominated homogeneous polynomials defined on $E$ having $μ$ as a Pietsch measure. As an application we fill the gap in the proofs of some results of concerning Pietsch-type factorization of dominated polynomials.

math.FA↗