Approximation and then-Berezin transform of operators on the Bergman space

Hilbert spaces of continuous, differentiable or analytic functions (46E20) Linear operator approximation theory (47A58) Toeplitz operators, Hankel operators, Wiener-Hopf operators (47B35) Linear operators in algebras (47C05) Algebras of specific types of operators (Toeplitz, integral, pseudodifferential, etc.) (47L80)

Review text

Let \(L^2_a\) be the Bergman space of all holomorphic functions in \(L^2=L^2(D,dA)\), where \(D\) is the unit disc and \(dA\) the normalized Lebesgue measure, and \(\mathcal L(L^2_a)\) the algebra of all bounded linear operators on~\(L^2_a\). For \(\phi\in L^\infty=L^\infty(D,dA)\), the Toeplitz operator \(T_\phi\in\mathcal L(L^2_a)\) is defined by \(T_\phi f =P(\phi f)\), where \(P:L^2\to L^2_a\) is the orthogonal projection. The Toeplitz algebra \(\mathcal T(L^\infty)\) is the \(C^*\)-subalgebra of \(\mathcal L(L^2_a)\) generated by \(\{T_\phi:\;\phi\in L^\infty\}\). For each \(z\in D\) and \(n,j\geq0\), consider the function \(k_{n,j,z}(x):=x^j/ (1-\overline zx)^{n+2}\). (Note that \(k_{0,0,z}(x)\) is precisely the reproducing kernel of~\(L^2_a\).) For \(S\in\mathcal L(L^2_a)\), the \(n\)-Berezin transform of \(S\) is the function on \(D\) defined~by \[ B_n(S)(z):=(n+1)(1-| z| ^2)^{n+2} \sum_{j=0}^n \binom nj (-1)^j \langle S k_{n,j,z},k_{n,j,z}\rangle. \] These transforms were introduced by the author in his earlier paper [Rev. Mat. Iberoam. 20, No. 2, 563--610 (2004; Zbl 1057.32005)] and used to study the structure of algebras generated by various classes of Toeplitz operators. Also, for \(n=0\), \(B_0(S)\)~coincides with the ordinary Berezin transform of \(S\), while for \(S=T_\phi\), \(B_n(T_\phi)\) coincides with the ordinary weighted Berezin transform of the function \(\phi\) [see, e.g., \textit{K. Zhu}, ``Operator theory in function spaces (Pure and Applied Mathematics 139, Marcel Dekker Inc., New York etc.) (1990; Zbl 0706.47019); \textit{S. Axler} and \textit{D. Zheng}, Indiana Univ. Math. J. 47, No. 2, 387--400 (1998; Zbl 0914.47029); \textit{P.~Ahern, M.~Flores} and \textit{W.~Rudin}, J. Funct. Anal. 111, No. 2, 380--397 (1993; Zbl 0771.32006)]. In the paper under review, the author first establishes a number of results concerning the approximation of \(S\) by \(T_{B_n(S)}\) as \(n\to\infty\). For~instance, \(T_{B_n(S)}\to S\) in the weak operator topology if and only if \(\sup_k\| T_{B_k(S)}\| <\infty\); \(T_{B_n(S)}\to S\) in norm if there exists \(p>4\) such that \(\sup_k \sup_{z\in D}(\| T_{B_k(S)\circ\phi_z}1\| _p + \| T^*_{B_k(S) \circ\phi_z}1\| _p)<\infty\), where \(\phi_z(x):=(z-x)/(1-\overline z x)\); and \(S=T_\phi\) for some \(\phi\in L^\infty\) if and only if \(\sup_k\| B_k(S)\| _\infty<\infty\). These results are then used to show that if \(S\) is radial (i.e., diagonal with respect to the standard basis \(\{z^n\}_{n\geq0}\) of~\(L^2_a\)), then \(S\) belongs to \(\mathcal T(L^\infty)\) if and only if \(T_{B_n(S)}\to S\) in operator norm. As~a corollary, it follows that a radial operator \(S\) is compact if and only if \(S\in\mathcal T(L^\infty)\) and \(B_0(S)\) vanishes at the boundary of~\(D\); and that there exists an operator \(S\in\mathcal L(L^2_a)\) whose commutator with \(T_z\) is compact but which does not belong to \(\mathcal T(L^\infty)\). The~latter answers a question of the reviewer from 1991. It~would be interesting to know if \(T_{B_n(S)}\to S\) even for arbitrary (i.e., not necessarily radial) \(S\) in \(\mathcal T(L^\infty)\).

Recommendations

Cited in

(21)

This page was built for publication: Approximation and then-Berezin transform of operators on the Bergman space

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4677656)