On the vertex operator representation of Lie algebras of matrices (Q2075926)

The very interesting paper under review can be seen as a finite type version of a celebrated result in representation theory, due to \textit{E. Date} et al. [J. Phys. Soc. Japan 50, 3806--3812 (1981; Zbl 0571.35099)]. The latter supplied a precise description of the ring \(\mathbf{C}[x_1,x_2,\ldots]\) in infinitely many indeterminates as a representation of the Lie algebra of the complex valued matrices of infinite size with finitely many non-zero diagonals. The DJKM representation is based on the fact that the so-called \textit{Fermionic Fock Space} can be seen as the fundamental representation of the infinite Lie algebra \(gl_\infty(\mathbb{C})\). From a representation theoretical point of view, the \textit{Fermionic Fock space} can be roughly thought of an infinite wedge power of the vector space \(\mathbb{C}[X^{-1}, X]\) of Laurent polynomials in the indeterminate \(X\). The natural question the authors asked themselves is how the DJKM picture can be detected already for polynomial rings in finitely many indeterminates. To explain what and how they do let us walk a few steps backward, to render more precisely the feeling of their result. To begin with, le \(K[X]\) be the ring of polynomials in one indeterminate over a field \(K\) of characteristic zero. Let \(B_r:=K[x_1,\ldots,x_r]\) be the \(K\)-algebra of polynomials in the indeterminates \(\mathbf{x}:=(x_1,\ldots,x_r)\). It is easy to convince oneself that \(B_r\) is isomorphic to the \(r\)-th exterior power \(\bigwedge^rK[X]\). Although a number of mathematicians like to explain this fact as a special case of some sophisticated \textit{Geometric Satake Correspondence}, the naive reason is that both spaces possess a basis parametrized by all the partitions of length at most \(r\). It is also easy to see that, as the \(r\)-th exterior power of \(K^n\) is a representation of the Lie algebra \(gl_n(K)\) of the \(K\)-valued \(n\times n\) square matrices, then \(\bigwedge^rK[X]\) turns into a representation of the Lie algebra \(gl_\infty(K)\) of the \(K\)-valued matrices \((a_{ij})_{i,j\geq 0}\) whose entries are all zero but finitely many. The isomorphism \(B_r\rightarrow \bigwedge^rK[X]\) then makes \(B_r\) itself into a representation of \(gl_\infty(K)\). The main result of the paper under review, Theorem 4.11, consists in determining what the authors call the \textit{generating formal power series} \(\mathcal{E}(z,w^{-1}, t_1,\ldots, t_r)\), which describes \(B_r\) as a representation of the Lie algebra \(gl_\infty(K)\). This means the following. Recall that the elementary matrices \(E_{i,j}\) with all entries zero but \(1\) in position \((i,j)\) form a basis of \(gl_\infty(K)\) and that, interpreting \(B_r\) as the ring of symmetric polynomials in \(r\) indeterminates, it possesse a basis of Schur determinants \(\Delta_{\lambda}\) constructed out of the complete symmetric polynomials, parametrized by partitions of length at most \(r\). If \(\mathbf{t}_r:=(t_1,\ldots,t_r)\) is an \(r\)-tuple of formal variables, denote by \(s_{\lambda}(\mathbf{t}_r)\) the Schur polynomials in the indeterminates \(\mathbf{t}_r\), like at p. 40 of the book by \textit{I. G. Macdonald} [Symmetric functions and Hall polynomials. With contributions by A. V. Zelevinsky. Reprint of the 1998 2nd edition. Oxford: Oxford University Press (2015; Zbl 1332.05002)]. The determination of the generating formal power series of the aforementioned Theorem 4.11 heavily relies on the techniques introduced in the 2005 reviewer's paper [\textit{L. Gatto}, Asian J. Math. 9, No. 3, 315--322 (2005; Zbl 1099.14045)] and substantially improves results by \textit{L. Gatto} and \textit{P. Salehyan} [Commun. Algebra 48, No. 1, 274--290 (2020; Zbl 1442.14156)]. How does this relate with DJKM work? This is widely discussed in the last section of the paper under review. In a nutshell, taking a suitable limit \(r\to\infty\), one obtains an isomorphism from the ring \(B:=B_\infty\) to the charge zero vector subspace \(\mathcal{F}_0\) of the \textit{fermionic Fock space} \(\mathcal{F}:=\bigwedge^{\infty/2}K[X^{-1}, X]\). There is a vector space isomorphism of \(\mathcal{F}_0\) with the ring \(B\) of polynomials in infinitely many indeterminates \((x_1,x_2,\ldots)\) (the \textit{bosonic Fock space}), because both spaces possess a basis parametrized by partitions. The literature (see, e.g., [\textit{V. G. Kac} et al., Highest weight representations of infinite dimensional Lie algebras. World Scientific (2013)]) often refers to this isomorphism as the \textit{boson-fermion correspondence.} As a consequence, \(B\) is made into a representation of the Lie algebra \(gl(\infty)\), induced by the natural one living on the infinite wedge power. Moreover the vertex operators occurring in the DJKM representation are nothing but an infinite dimensional version of what the authors call \textit{Schubert derivations}, borrowing the own reviewer terminology, which, as the name suggest, are devices useful to cope with Schubert Calculus. The paper concludes itself with an essential but comprehensive reference list.