Alcove path model for \(B(\infty)\) (Q2001423)

This paper is a contribution to the combinatorial theory of crystals for representations of quantum groups. Let \(\mathfrak{g}\) be a symmetrisable Kac-Moody algebra, and \(U_q(\mathfrak{g})\) its quantized enveloping algebra. A crystal for a representation of \(U_q(\mathfrak g)\) is a labelled directed graph which is derived by taking the limit \(q\to0\), and encodes much of the structure of the representation. The theory of crystals was introduced by \textit{M. Kashiwara} [Commun. Math. Phys. 133, No. 2, 249--260 (1990; Zbl 0724.17009); Duke Math. J. 63, No. 2, 465--516 (1991; Zbl 0739.17005)], and a central theme is to find combinatorial constructions of crystals in various cases. The aim in this paper is to give a construction of the crystal \(B(\infty)\); this is the crystal for the lower half \(U^-_q(\mathfrak g)\) of \(U_q(\mathfrak g)\). This is done starting from the Lenart-Postnikov alcove path model for crystals \(B(\lambda)\) of highest-weight representations. In this model, vertices of the crystal are given as finite sequences of positive roots for \(\mathfrak g\). In the present paper, a construction for \(B(\infty)\) is given by taking a direct limit as \(\lambda\to\infty\). This involves taking infinite sequences of positive roots. The central theorem of the paper is that the crystal constructed really is \(B(\infty)\). The final section considers the connection with the Littelmann path model (or rather, its infinite analogue given by \textit{B. Li} and \textit{H. Zhang} [Front. Math. China 6, No. 4, 689--706 (2011; Zbl 1282.17018)]). The paper is really nicely written, giving a clear introduction to the alcove path model, and then setting out the construction clearly. This is a paper that should be read by anyone interested in combinatorial constructions of crystals.