\(q\)-identities and affinized projective varieties. II: Flag varieties (Q1976012)

From MaRDI portal
Revision as of 16:39, 1 February 2024 by Import240129110113 (talk | contribs) (Added link to MaRDI item.)
scientific article
Language Label Description Also known as
English
\(q\)-identities and affinized projective varieties. II: Flag varieties
scientific article

    Statements

    \(q\)-identities and affinized projective varieties. II: Flag varieties (English)
    0 references
    0 references
    0 references
    7 December 2000
    0 references
    For part I of this paper see \textit{P. Bouwknegt}, ibid. 641-661 (2000; see the preceding review Zbl 0961.14039); definitions are to be found in that review. The theory of affinized projective varieties and their associated Hilbert series as developed in part I of this paper is applied to affinized flag varieties. The varieties \(V\) of part I are now of the form \(V=G/B\), where \(G\) is a (complex) Lie group with Lie algebra \(\text{Lie}(G)=\mathfrak g\), and where \(B\) is a Borel subgroup of \(G\). Besides the grading by multidegree \(M=(M_1,M_2,\ldots,M_{\ell})\), \({\mathcal S}(V)\) also has a grading by \(\mathfrak g\)-weight \(\lambda\). The Hilbert function of \({\mathcal S}(V)\) is defined as the (formal) \(\mathfrak g\)-character \[ h_V(M)=\sum_{\lambda}\dim{\mathcal S}(V)_{\mathbf M}:\lambda e^{\lambda}. \] The following Lie algebras are studied in some detail: \(\mathfrak{sl}_n\), in particular \(n=2,3,4\), \(\mathfrak{so}_{2n+1}\). -- Let \(E={\mathbb C}^n\) and fix a sequence \(n\geq d_s>\ldots >d_2>d_1\geq 0\). The (partial) flag variety \(\text{Fl}^{d_1\ldots d_s}(E)\) for the Lie algebra \(\mathfrak{sl}_n\) is the set of flags \(\{E_s\subset\ldots \subset E_2\subset E_1\subset E\mid \text{codim}E_i=d_i,1\leq i\leq s\}\). The (partial) Hilbert series of \(\widehat{V}\) is defined by \[ h_{\widehat{V}}(M ;q)= \sum_{N\geq 0,\lambda}\dim{\mathcal S}(\widehat{V})_{(\mathbf M;N;\lambda)}q^Ne^{\lambda}. \] For the purpose of computing Hilbert series non-quadratic monomial ideals can always be reduced to quadratic monomial ideals by the introduction of additional variables. This makes the application of part I possible. In particular, the \(q\)-identity (see part I) \[ {q^{M_1M_2}\over{(q)_{M_1}(q)_{M_2}}}=\sum_{m\geq 0}(-1)^m{q^{{1\over 2}m(m-1)} \over{(q)_m(q)_{M_1-m}(q)_{M_2-m}}} \] plays a basic role. The cases \(\mathfrak{sl}_n\), \(n=2,3,4\), and \(\mathfrak{so}_{2n+1}\), \(n=2,3\), are studied in detail. All these constructions lead to the result that expressions of the form \((\prod_{i=1}^{\ell}(q)_{M_i})h_{\widehat{V}}(M:q)\) are \(q\)-deformations of a \(\mathfrak g\)-character. This may suggest a relation to the so-called modified Hall-Littlewood polynomials for finite dimensional simple Lie algebras f rank \(\ell\). A precise conjecture is stated.
    0 references
    \(q\)-identities
    0 references
    affinized projective variety
    0 references
    flag variety
    0 references
    Hilbert series
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references