Sampling recovery of functions with mixed smoothness
This interesting paper studies sampling recovery of functions with mixed smoothness. We refer to the paper for an indepth summary of previous results in this huge area of research. We state their main result.\N\NLet \(\Omega\) be a compact subset of \(\mathbb R^d\), \(d\geq 1\) with probability measure \(\mu\) on it. For \(1\leq p<\infty\), denote by \(L_p\) the space of functions \(g:\Omega\to \mathbb C\) with \(||g||_{L_p(\Omega,\mu)}:= \left(\int_{\Omega}|g|d\mu\right)^{1/p}<\infty.\) For \(r>0\) and \(x=(x_1,...,x_d)\), let \(F_r(x)=\prod_{j=1}^df_r(x_j)\) with \(f_r(x):=1+2\sum_{k=1}^{\infty}k^{-r}\cos(kx-\frac{r\pi}{2})\). Now for \(1<q\leq 2\), define the smoothness class \(W_r^q:=\left\{f:f=\psi*F_r,\, ||\psi||_q\leq 1\right\}\) where \((\psi*F_r)(x):=(2\pi)^{-d}\int_{\tau^d}\psi(y)F_r(x-y)dy\) and \(\tau^d\) is the d-dimensional torus. With the notation above and \(1<q\leq 2\leq p<\infty\), let \(P_m(W_r^q,L_p(\tau^d)):=\inf_{\eta}\inf_{M}\sup_{f\in W_r^q}||f-M(f(\eta_1),..., f(\eta_m)||_p\) where \(\inf_{M}\) is over all maps \(M:\mathbb C^m\to L_p(\Omega,\mu)\) and for \(\inf_{\eta}\), \(\eta\) ranges over all subsets \((\eta_1,...,\eta_m)\) of \(m\) points in \(\Omega\).\N\NThe authors prove: With the notation above, there exist two absolute positive constants \(c:=c(r,d,p,q\)) and \(C:=C(r,d,p,q)\) so that\N\[\NP_m(W_r^q,L_p(\tau^d))\leq Cv^{-r+1/q-1/p}(\log v)^{(d-1)(r+1-2/q)}\N\]\Nfor every \(v\in \mathcal{N}\) and \(m\geq cv((\log (2v))^3)\). \N\NThe paper is well written with a good set of references.
- A new upper bound for sampling numbers
- A sharp upper bound for sampling numbers in \(L_2\)
- An Improved Estimate in the Restricted Isometry Problem
- Bounds on Kolmogorov widths and sampling recovery for classes with small mixed smoothness
- Constructive sparse trigonometric approximation and other problems for functions with mixed smoothness
- Constructive sparse trigonometric approximation for functions with small mixed smoothness
- Function values are enough for \(L_2\)-approximation
- Function values are enough for \(L_2\)-approximation. II
- Greedy algorithms in Banach spaces
- Greedy approximation
- scientific article; zbMATH DE number 43504 (Why is no real title available?)
- scientific article; zbMATH DE number 193625 (Why is no real title available?)
- scientific article; zbMATH DE number 3321507 (Why is no real title available?)
- Hyperbolic cross approximation. Lecture notes given at the courses on constructive approximation and harmonic analysis, Barcelona, Spain, May 30 -- June 3, 2016
- Multivariate approximation
- On approximate recovery of functions with bounded mixed derivative
- On optimal recovery in L₂
- Random points are good for universal discretization
- Sampling numbers of smoothness classes via \(\ell^1\)-minimization
- Sampling recovery on function classes with a structural condition
- Sparse sampling recovery in integral norms on some function classes
- The restricted isometry property of subsampled Fourier matrices
This page was built for publication: Sampling recovery of functions with mixed smoothness
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6946592)
