Nonlinear weighted best simultaneous approximation in Banach spaces

Simultaneous approximation (41A28) Best approximation, Chebyshev systems (41A50) Abstract approximation theory (approximation in normed linear spaces and other abstract spaces) (41A65)

Review text

Let \(\mathbb{R}^{\infty}\) be a Banach space of real sequences with a monotone norm \(\| .\| _A\) such that \[ \lim_{\nu \rightarrow \infty} \| (0,\dots,0,\lambda_{\nu},\lambda_{\nu+1},\dots) \| _A =0. \] Let \(\lambda_{\nu} = \frac{1}{(\nu^2 \| e^\nu \| _A)}\). Let \({\mathcal F} = \{\widehat{x}=(x_{\nu}): (\lambda_{\nu} \| x_{\nu}\| ) \in\mathbb R^{\infty}\}\), equipped with the norm \(\| \widehat{x}\| = \| (\lambda_{\nu} \| x_{\nu}\| )\| _A\). Let \({\mathcal F}_B = \{\widehat{x}=(x_{\nu}): (\| x_{\nu}\| )\) is bounded\}. Let \(X\) be a real or complex Banach space. A set \(G \subset X\) is said to be a BS-sun, if for any \(\widehat{x} \in {\mathcal F}_B\), \(g_0 \in P_G(\widehat{x})\Rightarrow g_0 \in P_G(\widehat{x}_{\alpha})\), where \(\widehat{x}_{\alpha} = g_0 + \alpha (\widehat{x}-g_0)\) for all \(\alpha>0\). One of the main results of this paper is to characterize BS-sun' as those for which \(g_0 \in P_G(\widehat{x})\) if and only if \(\max\{(a^\ast,(\Re\, \lambda_{\nu} f_{\nu}(g_0-g)))_A: (a^\ast,\widehat{f})\in E_{\widehat{x}-g_0}\} \geq 0\) for all \(g \in G\). Here \(E_{\widehat{x}-g_0}\) is an appropriate extremal set.

Recommendations

Cites work

Cited in

(7)

This page was built for publication: Nonlinear weighted best simultaneous approximation in Banach spaces

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2473822)