The Kadison-Singer problem (Q254938)

The paper under review is a survey paper, written in French, concerning the famous Kadison-Singer problem. This problem, which was proposed in [Am. J. Math. 81, 383--400 (1959; Zbl 0086.09704)] by \textit{R. V. Kadison} and \textit{I. M. Singer}, asks whether every pure state on the algebra \(\mathcal{D}(\ell^{2})\) of diagonal operators on the Hilbert space \(\ell^{2}\) can be uniquely extended to a state on the algebra \(\mathcal{B} (\ell^{2})\) of all bounded operators on \(\ell^{2}\). This problem admits a striking variety of equivalent formulations, pertaining to domains which are sometimes quite far away from the initial context of the Kadison-Singer problem, namely \(C^{*}\)-algebras. The problem was solved in 2013 by \textit{A. W. Marcus} et al. [Ann. Math. (2) 182, No. 1, 327--350 (2015; Zbl 1332.46056), \url{arXiv:1306.3969}], using relatively elementary methods in an extremely ingenious way. The aim of this survey is not to present the proof of [loc.\,cit.]\ -- the reader is referred instead to some other references, like for instance [\textit{A. Valette}, Astérisque 367--368, 451--476, Exp. No. 1088 (2015; Zbl 1356.46002)]. The author rather takes the reader to a tour into the Kadison-Singer problem itself and its various reformulations, leaving him/her at the door of the solution at the end. The text, being issued from notes given at a spring school, is aimed at students, and written in an extremely clear and challenging way. Tools on \(C^{*}\)-algebras, filters, ultrafilters, etc., necessary in order to understand the Kadison-Singer problem are presented in detail. Compressibility, paving of bounded operators on \(\ell^{2}\) and the Dixmier property are dealt with, and some non-trivial equivalent reformulations of the Kadison-Singer conjecture are given. Relations with the Feichtinger conjecture and the Bourgain-Tzafriri conjecture are also presented.