On two equivalent dilation theorems in VH-spaces (Q371693): Difference between revisions

A well-known theorem of Sz.-Nagy states that a bounded positive-definite function, defined on a *-semigroup and taking values in the bounded operators on a Hilbert space, can be dilated to a *-representation (this appeared in the appendix of [\textit{F. Riesz} and \textit{B. Sz.-Nagy}, Functional analysis. London-Glasgow: Blackie \& Son, Ltd (1956; Zbl 0070.10902); Reprint (1990; Zbl 0732.47001)]; see also a less general result in Section I.7 of [\textit{B. Sz. -Nagy} et al., Harmonic analysis of operators on Hilbert space. New York: Springer (2010; Zbl 1234.47001)]). This general dilation theorem encapsulates many dilation theorems about operators on Hilbert space. It is known that Stinespring's dilation theorem for completely positive maps on C*-algebras [\textit{W. F. Stinespring}, Proc. Am. Math. Soc. 6, 211--216 (1955; Zbl 0064.36703)] can be proved using Sz.-Nagy's dilation theorem, and \textit{F. H. Szafraniec} [in: Functions, series, operators, Proc. int. Conf., Budapest 1980, Vol. II, Colloq. Math. Soc. János Bolai 35, 1165--1169 (1983; Zbl 0543.47007)] showed that one can prove Sz.-Nagy's theorem using Stinespring's theorem. Though it seems that Sz.-Nagy's dilation theorem is the most general possible dilation theorem, in [Proc. Lond. Math. Soc. (3) 15, 373--384 (1965; Zbl 0138.39102)], \textit{R. M. Loynes} generalized this theorem to the setting where the positive definite function takes values in a VH-space, that is, a topological vector space where the topology is generated by a suitable topological-vector-space-valued inner product. The paper under review achieves three goals. First, it gives a nice review of VH-spaces and of Loynes's theorem, while improving the theorem in a nontrivial way by removing some assumptions. Second, a version of Stinespring's theorem appropriate to VH-spaces is proved (in two ways: from the generalized Sz.-Nagy dilation theorem and directly, mimicking Stinespring's original proof). Finally, Loynes's generalization of the Sz.-Nagy dilation theorem is deduced from the generalized Stinepring theorem. The paper is well written and reader friendly. Emphasis is put where it should: on overcoming the obstructions arising from the VH-space anomalies. It is also nice that the paper contains several different examples of VH-spaces, so that the reader not acquainted with them can see that there is substance in this generalization.