Schur algebras for the alternating group and Koszul duality (Q2189535)

The main object of study in the paper under review is the alternating Schur algebra \(AS_F(n,d)\), which is the commutant of the action of the alternating group \(A_d\) on the \(d\)-fold tensor power of an \(n\)-dimensional \(F\)-vector space. The main motivation to study this object is the link with the Koszul duality functor on the category of homogeneous polynomial representations of \(\operatorname{GL}_n(\mathbb{C})\) of degree \(d\). In the following we assume that \(F\) has characteristic different from \(2\). The first main result gives a basis for \(AS_F(n,d)\) and the structure constants with respect to this basis. This basis has a combinatorial flavour and is described in terms of bipartite simple graphs. In the next part of the paper, the authors introduce the term abstract Koszul duality. Suppose that \(AS = S\oplus S^-\) is a \(\mathbb{Z}/2\mathbb{Z}\)-graded algebra. Then the abstract Koszul duality associated with \(AS\) is a functor on the category of \(S\)-modules defined by \[ D(M) = S^-\otimes_S M. \] The authors first study some properties of the abstract Koszul duality in the general case and they prove that the category of \(AS\)-modules is the same as the category of pairs \((M,\theta_M)\), where \(M\) is an \(S\)-module and \(\theta_M :M \to M\) is a homomorphism compatible (in the sense that it gives rise to a certain natural commutative diagram) with the multiplication \(S^-\otimes_S S^- \to S\). Then they restrict their study of the abstract Koszul duality to the alternating Schur algebra \(AS_F(n,d)\). The authors prove that when \(n \geq d\) the evaluation at \(F^n\) of the Koszul duality functor of \textit{H. Krause} [Compos. Math. 149, No. 6, 996--1018 (2013; Zbl 1293.20046)] is isomorphic to the abstract Koszul duality. Moreover with the additional assumption that the characteristic of \(F\) is either \(0\) or greater than \(d\), the abstract Koszul duality is an equivalence. Finally, the authors show that when \(n < d\) the situation is different. They prove that the derived Koszul duality is not an equivalence in this case. Their methodology relies on their combinatorial description of the structure constants of \(AS_F(n,d)\).