\(L\)-functions and theta correspondence for classical groups (Q2247932)

In this paper the author discusses two very important problems in the theory of automorphic forms, local representation theory and theta correspondences. The first problem is the definition of the local factors for the representations of the classical groups (in the wider sense), i.e., the reconciliation of the definitions of local factors by \textit{I. I. Piatetski-Shapiro} and \textit{S. Rallis} [Proc. Natl. Acad. Sci. USA 83, 4589--4593 (1986; Zbl 0599.12012)] with those of \textit{E. M. Lapid} and \textit{S. Rallis} [in: Automorphic representations, \(L\)-functions and applications. Columbus, USA, 2003. Berlin: de Gruyter 309--359 (2005; Zbl 1188.11023)]. The second problem is the characterization of the non-vanishing of the global theta lift in terms of the non-vanishing of the local lifts plus a condition on the global \(L\)-function. To obtain these results, the author uses the doubling method of Piatetski-Shapiro and Rallis (generalizing the work of \textit{R. Godement} and \textit{H. Jacquet} [Lect. Notes Math. 260 (1972; Zbl 0244.12011)] on \(\mathrm{GL}(n)\)) and carefully discusses all the special cases of the groups he considers (besides the split classical groups, he discusses unitary and quaternionic groups together with the general linear and division algebra cases). Many results for the split cases, were, in some form, already known, but the author makes them very precise, and covers the rest of the cases. Considering the first problem, the main difficulty is to show that the two definitions of the \(L\)-functions mentioned above, agree. Namely, Piatetski-Shapiro and Rallis define \(L\)-factors as the greatest common divisor of a family of zeta integrals. On the other hand Lapid and Rallis first define the gamma factor as a constant that appears in the functional equation satisfied by these zeta integrals, and then, first in tempered situation, and then in for general representations, define \(L\)-factors from gamma factors. Previous results on the non-vanishing of global theta lifts provided mainly sufficient conditions for the non-vanishing in terms of local lifts and partial \(L\)-functions (except in some small rank cases e.g. in the work of Waldspurger); but the author gives complete answers for most cases in terms of the completed \(L\)-functions. For a simpler exposition, the author describes in the introduction the main two results in the case of orthogonal groups.