Poincaré series for tensor invariants and the McKay correspondence (Q908062): Difference between revisions

Let \(G\) be a finite group and \(\Lambda(G)\) be the set of irreducible complex representations of \(G\). Associated to a fixed finite-dimensional representation \(V\), are the numbers \(m_k^\lambda\), which are the multiplicity of \(\lambda\in \Lambda(G)\) in \(V^{\otimes k}\), \(A=(a_{\lambda \, \mu})\), the adjacency matrix of the representation graph \(R_V(G)\), and the Bratteli diagram which is the infinite graph with vertices labeled by the elements of \(\Lambda(G)\) on level \(k\), keeping tracks of finite steps walks on \(R_V(G)\). The Poincaré series for \(\lambda\in \Lambda(G)\) is defined by \[ m^{\lambda}(t)=\sum_{k=0}^\infty m_k^\lambda t^k. \] The main result of the paper under review is as follows: suppose that \(G\) acts faithfully on \(V\), and \(V^* \cong V\) as \(G\)-representations. Let \(M_\mu\) be the matrix \(I - tA\) with the column indexed by \(\mu\) replaced by \((1\; 0\; 0 \dots 0)^T\). Then \[ m^\mu(t)=\frac{\det M_\mu}{\det(I-tA)}. \] In the special case where \(G\) is a finite subgroup of \(\mathrm{SU}(2)\), and \(V=\mathbb{C}^2\) is its natural representation, then \[ m^0(t)=\frac{\det(I-t\hat{A})}{\det(I-tA)}, \] where \(0\in \Lambda(G)\) is the trivial representation, and \(\hat{A}\) is the adjacency matrix of the finite Dynkin diagram obtained by removing the affine node. In the continuation, the paper under review finds further explicit formulas for \(m^0(t)\) and express the results in terms of Chebyshev polynomials, and the Bratteli diagrams.