The geometry of unitary 2-representations of finite groups and their 2-characters (Q633796): Difference between revisions

For a Lie group \(G\), geometric quantization in the \(G\)-equivariant context is based on an equivalence of (complexified Grothendieck groups of) categories between the category of unitary representations of \(G\) and the category of \(G\)-equivariant line bundles. Moreover, there is a correspondence between the character of a representation of \(G\) and the suitably-defined geometric character of its associated equivariant line bundle. Extended topological quantum field theory (ETQFT) would associate a 2-category of 2-repre\-sen\-ta\-tions to a point, and so we would want to categorify the above geometric picture. The paper under review achieves this goal for a concrete toy model. The setting of the paper is a 2-Hilbert space in the sense of \textit{J. C. Baez} [Adv. Math. 127, No.~2, 125--189 (1997; Zbl 0896.18001)], being acted on unitarily and coherently by a finite group \(G\). The 2-category which these 2-Hilbert spaces constitute is semisimple, meaning that all relevant geometric structures are discrete. Theorem~9.1 establishes the equivalence of this 2-category with the 2-category of \(G\)-equivariant gerbes equipped with a measure. Similar ideas appear also in \textit{J. Elgueta} [Adv. Math. 213, No.~1, 53--92 (2007; Zbl 1118.18003)] and in \textit{J. W. Barrett} and \textit{M. Mackaay} [Theory Appl. Categ. 16, 529--557 (2006; Zbl 1108.18003)] in a different language; but the language of gerbes is more geometric. The heart of the paper is the categorification of the correspondence between characters of representations and geometric characters. Both sides of this correspondence must be categorified. Using string diagrams, group characters are categorified to 2-characters of 2-representations. In a similar vein to the work of \textit{S. Willerton} [Algebr. Geom. Topol. 8, No.~3, 1419--1457 (2008; Zbl 1154.57029)], a geometric character of an equivariant gerbe is defined as the space of sections over its fixed points (push-forward of the transgression map). Theorem 10.1 is that (in this toy setting) 2-characters are unitarily naturally isomorphic to their associated geometric characters. Corollary 10.2 is that the complexified 2-character functor is unitarily fully faithful, which categorifies the fact that the (ordinary) character is a unitary map from the complexified Grothendieck group of unitary representations to the space of class functions. \textit{N. Ganter} and \textit{M. Kapranov} [Adv. Math. 217, No.~5, 2268--2300 (2008; Zbl 1136.18001)] independently defined 2-characters of 2-representations, motivated by equivariant homotopy theory, and there is some overlap in these sections.