Greedy bases in variable Lebesgue spaces (Q258067)

From MaRDI portal





scientific article; zbMATH DE number 6557689
Language Label Description Also known as
default for all languages
No label defined
    English
    Greedy bases in variable Lebesgue spaces
    scientific article; zbMATH DE number 6557689

      Statements

      Greedy bases in variable Lebesgue spaces (English)
      0 references
      0 references
      0 references
      0 references
      17 March 2016
      0 references
      Let \(X\) be a Banach space and let \(\{b_j\}\) be a Schauder basis for \(X\). Denote by \(\Sigma_N\) the set of all \(x\in X\) with at most \(N\) non-zero coefficients in their basis representation \(x=\sum_{j=1}^{\infty}\lambda_jb_j\). Given \(x\in X\) reorder the basis elements so that \[ \|\lambda_{j_1}b_{j_1}\|\geq\|\lambda_{j_2}b_{j_2}\|\geq\|\lambda_{j_3}b_{j_3}\|\geq\dots \] and then we set \[ G_N(x)=\sum_{k=1}^N\lambda_{j_k}b_{j_k}. \] A basis is greedy if there exists \(C>1\) such that, for all \(x\in X\) and \(N\), \[ \|x - G_N(x)\|\leq C\inf_{y\in\Sigma_N}\|x - y\|. \] A basis is democratic if given any two index sets \(\Gamma\), \(\Gamma'\), \(\mathrm{card}(\Gamma)=\mathrm{card}(\Gamma')\), \[ \left\|\sum_{j\in\Gamma}\frac{b_j}{\|b_j\|}\right\|\approx\left\|\sum_{j\in\Gamma'}\frac{b_j}{\|b_j\|}\right\|. \] \textit{S. V. Konyagin} and \textit{V. N. Temlyakov} [East J. Approx. 5, No. 3, 365--379 (1999; Zbl 1084.46509)] characterized greedy bases as those which are unconditional and democratic. Given an exponent function \(p(\cdot)\), let \(L^{p(\cdot)}(\mathbb{R}^n)\) denote the variable Lebesgue space on \(\mathbb{R}^n\). Many wavelet bases form unconditional bases on the variable Lebesgue spaces; however, unless \(p(\cdot)=p\) is constant, they are never democratic in \(L^{p(\cdot)}(\mathbb{R}^n)\). The authors quantify the failure of democracy by computing precisely the right and left democracy functions of admissible wavelet bases in \(L^{p(\cdot)}(\mathbb{R}^n)\). As an application Lebesgue typy inequalities for these wavelet bases are obtained. It is shown also that this techniques can be easily modified to prove analogous results for weighted variable Lebesgue spaces and variable exponent Triebel-Lizorkin spaces.
      0 references
      greedy algorithm
      0 references
      non-linear approximation
      0 references
      wavelets
      0 references
      variable Lebesgue spaces
      0 references
      Lebesgue type estimates
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references