scientific article; zbMATH DE number 4103347

From MaRDI portal
Publication:3828192





Let \(G_ n=GL_ n({\mathbb{C}})\) and \(P_ n\) be the weight lattice of \(G_ n\) which is identified with \({\mathbb{Z}}^ n\) in the standard way. Let \(P^+_ n\) be the set of highest weights of irreducible \(G_ n\)- modules, i.e. \(P^+_ n=\{\lambda =(\lambda_ 1,...,\lambda_ n)\in P_ n:\) \(\lambda_ 1\geq \lambda_ 2\geq...\geq \lambda_ n\}\). If \(\lambda \in P^+_ n\) then we denote by \(V_{\lambda}\) the irreducible \(G_ n\)-module with the highest weight \(\lambda\). If \(0\leq k\leq n\) then \(G_ k\times G_{n-k}\) is identified with the subgroup \(\left( \begin{matrix} G_ k\\ 0\end{matrix} \begin{matrix} 0\\ G_{n-k}\end{matrix} \right)\) of \(G_ n\), and according to this we write \(\gamma \in P_{n- k}\) as \(\gamma =(\gamma_{k+1},\gamma_{k+2},...,\gamma_ n)\). We denote also \(V_{\lambda | \rho}=Hom_{G_ k}(V_{\rho},V_{\lambda}|_{G_ k})\) for all \(\lambda \in P^+_ n\), \(\rho \in P^+_ k\) (if \(k=0\) then we assume that \(\rho =\emptyset\) and \(V_{\lambda | \emptyset}=V_{\lambda}).\) Let \(\lambda =(\lambda_ 1,...,\lambda_ n)\in P_ n\), \(\rho =(\rho_ 1,...,\rho_ k)\in P_ k\). Then any set \(\Lambda =\{\lambda^{(j)}=(\lambda_ 1^{(j)},...,\lambda_ j^{(j)})\in P_ j\), \(k\leq j\leq n\}\) such that \(\lambda^{(k)}=\rho\), \(\lambda^{(n)}=\lambda\) and \(\lambda_ i^{(j+1)}\geq \lambda_ i^{(j)}\) for all \(1\leq i\leq j<n\) is called a truncated Gel'fand-Tsetlin semischeme of type \(\lambda\) \(| \rho\). The weight \(\gamma =(\gamma_{k+1},...,\gamma_ n)\in P_{n-k}\) defined by the equalities \(\rho_ 1+...+\rho_ k+\gamma_{k+1}+...+\gamma_ j=\lambda_ 1^{(j)}+...+\lambda_ j^{(j)}\) for all \(k<j\leq n\) is called the weight of the semischeme \(\Lambda\). The semischeme \(\Lambda\) is called scheme if \(\lambda_ i^{(j)}\geq \lambda_{i+1}^{(j+1)}\) for all \(1\leq i\leq j<n.\) The following two theorems are proved. Theorem 1. The multiplicity of the weight \(\gamma\) in a \(G_{n-k}\)-module \(V_{\lambda | \rho}\) is equal to the number of all truncated Gel'fand-Zetlin schemes of type \(\lambda\) \(| \rho\) and weight \(\gamma\). - Let \(\Lambda =\{\lambda^{(k)},...,\lambda^{(n)}\}\) be a truncated scheme. One calls \(\Lambda\) \(\nu\)-restricted for \(\nu \in P^+_{n-k}\) if one has \[ \sum_{1\leq h<i}(\lambda_ h^{(j+1)}-2\lambda_ h^{(j)}+\lambda_ h^{(j-1)})+\lambda_ i^{(j+1)}-\lambda_ i^{(j)}\leq \nu_ j-\nu_{j+1} \] for all i, j. Theorem 2. Let \(\lambda \in P^+_ n\), \(\rho \in P^+_ k\), \(\mu,\nu \in P^+_{n- k}\). Then the multiplicity of the \(G_{n-k}\)-module V in \(V_{\lambda | \rho}\otimes V_{\nu}\) equals the number of all \(\nu\)-restricted schemes of type \(\lambda\) \(| \rho\) and weight \(\mu\)-\(\nu\).











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3828192)