Embedding subspaces on L_p into _p^N, 0
Given an \(n\)-dimensional subspace \(X\) of \(L_p[0,1]\) and \(\varepsilon > 0\), what is the smallest integer \(N = N(n, \varepsilon)\) such that there is a subspace \(Y\) of \(l_p^N\) with \(d(X, Y) \leq 1 + \varepsilon\), where \(d\) is the Banach--Mazur distance? For \(p \geq 1\), the answer is known, except for possibly redundant \(\log\) factors (ignoring such factors, it is \(n\) for \(1 \leq p \leq 2\), and \(n^{p/2}\) for \(2 \leq p < \infty\)). For \(0 < p < 1\), the best known result is \(N(n, \varepsilon) \leq c(\varepsilon, p)n^2\), due to the second author [\textit{G. Schechtman}, Compos. Math. 61, 159--169 (1987; Zbl 0659.46021)]. Recently this was generalized by \textit{A.~Peña} [Math. Nachr. 189, 195--207 (1998; Zbl 0908.46011)] who showed that for any \(0 < p < 1\), any \(X\) as above admits a \(Y \subseteq l_p^N\) with \(N \leq C(p)n\) and \(d(X, Y) \leq C(p) (\log n)^{1/p}\). The problem of reducing the \(\log n\) factor to \(1 + \varepsilon\) was left open.NEWLINENEWLINENEWLINENEWLINEIn the present paper, the authors show that \(N(n, \varepsilon) \leq C(p, \varepsilon) n (\log n)^3\). Proofs for \(p \geq 1\) use Lewis' change of density result which previously was known only in that range [cf. \textit{D.~R. Lewis}, Stud. Math. 63, 207--212 (1978; Zbl 0406.46023); Mathematika 26, 18--29 (1979; Zbl 0438.46006)]. The authors extend Lewis' result to \(0 < p < \infty\) with a proof that is new even in the case \(p \geq 1\).NEWLINENEWLINETheir version of Lewis' theorem is that if \(X\) is an \(n\)-dimensional subspace of \(L_p(\Omega, \mu), 0 < p < \infty\), there is a basis \(f_1, \ldots, f_n\) of \(X\) satisfying NEWLINE\[NEWLINE \int \left( \sum_{i =1}^n f_i^2 \right)^{(p -2)/2} f_k f_l \, d\mu = \delta_kl, \quad 1 \leq k,\;l \leq n. NEWLINE\]
- scientific article; zbMATH DE number 775040
- Embedding Subspaces of L 1 into l N 1
- scientific article; zbMATH DE number 1574611
- A Note on the Embedding of Subspaces ofLp into ℓNr, 0 < r < 1, r ≦p < 2
- Embedding \(\ell_ p^ k\) in subspaces of \(L_ p\) for \(p>2\)
- scientific article; zbMATH DE number 860022
- scientific article; zbMATH DE number 4078003
- Very tight embeddings of subspaces of \(L_p\), \(1 \leq p < 2\), into \(\ell^n_p\)
- Tight embedding of subspaces of 𝐿_{𝑝} in ℓ_{𝑝}ⁿ for even 𝑝
- Subspaces of \(L_p\) that embed into \(L_p(\mu)\) with \(\mu\) finite
- Subspaces of L_p isometric to subspaces of _p
- Very tight embeddings of subspaces of \(L_p\), \(1 \leq p < 2\), into \(\ell^n_p\)
- On Dvoretzky's theorem for subspaces of \(L_p\)
- Marcinkiewicz-type discretization of L^p-norms under the Nikolskii-type inequality assumption
- Sampling discretization and related problems
- On extremal sections of subspaces of L_p
- An isomorphic version of the Busemann-Petty problem for arbitrary measures
- Codimension of subspaces of \(L_ p(\mu)\) for \(p<1\)
- Embeddability of \(\ell_p\) and bases in Lipschitz free \(p\)-spaces for \(0 < p \leq 1\)
- Different degrees of non-compactness for optimal Sobolev embeddings
- Slicing inequalities for subspaces of L_p
- Tight embedding of subspaces of 𝐿_{𝑝} in ℓ_{𝑝}ⁿ for even 𝑝
- RESTRICTED INVERTIBILITY AND THE BANACH–MAZUR DISTANCE TO THE CUBE
- Embedding Subspaces of L 1 into l N 1
- A Note on the Embedding of Subspaces ofLp into ℓNr, 0 < r < 1, r ≦p < 2
- Subspace embeddings for the L 1 -norm with applications
- Some improved bounds in sampling discretization of integral norms
- Weighted Minkowski’s existence theorem and projection bodies
- Monotonicity of functionals associated to product measures via their Fourier transform and applications
- Online Lewis weight sampling
- Sampling discretization in Orlicz spaces
- On good \({\mathcal L}_ p\) subspaces of \(\ell _ p\)
This page was built for publication: Embedding subspaces on \(L_p\) into \(\ell_p^N, 0<p<1\)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2747574)
