The Dixmier trace and other traces (Q330035): Difference between revisions

In [C. R. Acad.\ Sci., Paris, Sér.~A 262, 1107--1108 (1966; Zbl 0141.12902)], \textit{J.~Dixmier} constructed a nonzero trace on \(L(H)\), the von Neumann algebra of bounded linear operators on a Hilbert space~\(H\), that vanishes for all compact operators. For this reason, such a trace is called singular. In the present paper, the author first presents the necessary prerequisites for constructing such a trace (among them, the ordinary trace of nuclear operators, the Courant-Fischer variational principle, Banach limits and the Hahn-Banach theorem, invariant means) and then continues with a detailed desription of Dixmier's trace including the ``delicate'' proof of its additivity. The final section is devoted to a discussion of the importance of singular traces in noncommutative geometry as first envisioned by {A.~Connes} (spectral triples, Fredholm modules). Reviewer's remark: The reader might also like to study the papers [\textit{A. Pietsch}, Acta Comment. Univ. Tartu. Math. 18, No. 1, 51--64 (2014; Zbl 1300.47005)] and [\textit{A. L. Carey} and \textit{F. A. Sukochev}, Russ. Math. Surv. 61, No. 6, 1039--1099 (2006); translation from Usp. Mat. Nauk 61, No. 6, 45--110 (2006; Zbl 1151.46053)] for origins, applications and generalisations of Dixmier's construction.