Rank and duality in representation theory (Q783808)

In the introduction, the authors review the theory of ``rank'' or ``size'' of representations of classical groups over finite fields and local fields. It also explain some motivation which involves the character ratios for an irreducible representation of a finite group. It is very readable and contains many historical facts. In the body of the article, several novel definitions of ranks (called \(U\)-rank, tensor rank, and asymptotic rank) for representations of classical groups over finite fields were introduced. For a representation of \(\mathrm{GL}_n\), the \(U\)-rank involves the largest rank among all \(U\)-isotropic components of the representation, as \(U\) varies all standard unipotent matrix subgroups of \(\mathrm{GL}_n\), while the tensor rank describes the exact integer \(k\), as the representation sits in \(k\)-fold tensor products of \(U\)-rank \(1\) irreducible representations. The asymptotic rank is another rank between them. A low \(U\)-rank representation is a kind of representation whose \(U\)-rank double is strictly smaller than \(n\). For other classical groups, these definitions were given based on the authors' recently works [Prog. Math. 323, 209--234 (2017; Zbl 1405.20036)]. It should be mentioned that the \(U\)-rank \(1\) irreducible representations of \(Sp\) consists of the four components of Weil representations. In this paper, the authors also propose several conjectures. One conjecture is the following: for \(\mathrm{GL}\) and \(\mathrm{Sp}\), the \(U\)-rank and tensor rank of a low \(U\)-rank representation are the same. To construct representations of a given rank, the authors develop a method called eta correspondence. This correspondence refines the celebrated theta or Howe correspondence by matching rank requirements. [\textit{R. Howe}, J.Am. Math. Soc. 2, No. 3, 535--552, (1989; Zbl 0716.22006); \textit{C. Mœglin} et al., Correspondances de Howe sur un corps \(p\)-adique. Berlin etc.: Springer-Verlag (1987; Zbl 0642.22002)]. In the finite field setting, it can provide some explicit and exact correspondences between representations of dual pairs, which were not proposed or exhibited before. ([\textit{B. Srinivasan}, Invent. Math. 51, 143--153 (1979; Zbl 0399.20040); \textit{J. Adams} and \textit{A. Moy}, Trans. Am. Math. Soc. 340, No. 1, 309--321 (1993; Zbl 0829.22025); \textit{ A.-M. Aubert} et al., Duke Math. J. 83, No. 2, 353-397 (1996; Zbl 0856.22027); \textit{R. Howe}, ``Invariant theory and duality for classical groups over finite fields with applications to their singular representation theory'', Preprint].