The Farrell-Hsiang method revisited (Q453311)

The paper deals with the Farrell-Jones conjecture which, given a group \(G\) and a ring \(R\), makes a statement on the \(K\)- and \(L\)-theory of the group ring \(RG\), namely that the latter essentially can be computed from homological data of the group \(G\) and the \(K\)- and \(L\)-theory of the ring \(R\). For further information on the Farrell-Jones conjecture we recommend the excellent survey article by \textit{W. Lück} and \textit{H. Reich} [in: E. M. Friedlander (ed.) et al., Handbook of \(K\)-theory. Vol. 1 and 2. Berlin: Springer. 703--842 (2005; Zbl 1120.19001)]. The Farrell-Jones conjecture for a given group most often is proved using methods of controlled topology, and one of the major ingredients then always is to find a tool for improving the relevant control data. With the so-called Farrell-Hsiang method, which was first used in [\textit{F. T. Farrell} and \textit{W. C. Hsiang}, Invent. Math. 45, 181--192 (1978; Zbl 0422.57019)], this improved control is gained by means of a transfer argument which is based on Frobenius induction. In the paper at hand the authors now derive group-theoretic, respectively group-geometric, properties of the group \(G\) which allow to run the Farrell-Hsiang argument for establishing the so-called Farrell-Jones conjecture relative to a given family of subgroups \({\mathcal F}\). They call a group having these properties a Farrell-Hsiang group for the family \({\mathcal F}\), and they then indeed show that the Farrell-Jones conjecture relative to \({\mathcal F}\) holds for such groups. This new concept of Farrell-Hsiang groups has meanwhile been used prominently in [\textit{A. Bartels} et al., J. Am. Math. Soc. 27, No. 2, 339--388 (2014; Zbl 1307.18012)], where the three authors show that the Farrell-Jones conjecture holds for virtual poly-cyclic groups and even for cocompact lattices in virtually connected Lie groups.