Rate of convergence of thresholding greedy algorithms (Q6569587)

This paper is a follow-up to the paper [\textit{V. N. Temlyakov}, ``On the rate of convergence of greedy algorithms'', Preprint, \url{arXiv:2304.06423}]. In that paper the author found the rate of convergence of some standard greedy algorithms with respect to arbitrary dictionaries. There he bounded the error of approximation by the product of both the norm of \(f\) and the \(A_1\)-norm of \(f\). This is an important feature of [loc. cit.], since typically only the \(A_1\)-norm of \(f\) is used. This paper is focused on a special class of dictionaries, namely, on various types of bases -- greedy bases, unconditional bases, and quasi-greedy bases. The author studies the rate of convergence of the classical Thresholding Greedy Algorithm with respect to bases in the style of [loc. cit.]. In particular, he proves that the given bounds for the trigonometric basis and Haar basis are optimal.