Optimization of active algorithms for recovery of monotonic functions from Hölder's class
Let \(H^ \alpha\) be the class of all functions \(f\in C([a,b])\) satisfying for \(0< \alpha\leq 1\) the condition \(| f(x')- f(x)|\leq | x'- x|^ \alpha\), \(x',x\in [a,b]\) and let \(H^ \alpha_ m\) be the set of monotonic functions in \(H^ \alpha\). The relationship of passive and active algorithms in problems of recovery of functions is considered. The author proves that bisection is optimal among all active algorithms in the problem of recovery of functions \(f\in H^ \alpha_ m\) with metric \(C\) and when \(0< \alpha<1\) it achieves the best order \(O(N^{- 1}\log_ 2 N)\) on the set \(H^ \alpha_ m\).
- On passive and active algorithms of reconstruction of functions
- Optimization of adaptive algorithms for the renewal of monotone functions from the class H^
- Generalized adaptive versus nonadaptive recovery from noisy information
- scientific article; zbMATH DE number 3928214 (Why is no real title available?)
- Optimal renewal of definite integrals of monotone functions from the class H ω
- Joint spectra of Toeplitz operators and optimal recovery of analytic functions
- Homogeneous algorithms and solvable problems on cones
- Approximation of functions: optimal sampling and complexity
This page was built for publication: Optimization of active algorithms for recovery of monotonic functions from Hölder's class
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1333246)
