A canonical form for nonderogatory matrices under unitary similarity (Q541948)

A square matrix is called nonderogatory if its characteristic and minimal polynomials coincide. The authors of the paper under review give canonical forms for nonderogatory matrices up to unitary similarity and, pairs of matrices up to similarity in which one matrix has distinct eigenvalues. They use Littlewood and Belitskii algorithm [\textit{G. Belitskii}, Normal forms in a space of matrices, in: V. A. Marchenko (ed.), Analysis in infinite-dimensional spaces and operator theory, Naukova Dumka, Kiev, 3--15 (in Russian) (1983)], [\textit{G. Belitskii}, Integral Equations Oper. Theory 38, No. 3, 251--283 (2000; Zbl 0971.65037)], [\textit{D. E. Littlewood}, J. Lond. Math. Soc. 28, 314--322 (1953; Zbl 0050.25103)] and Schur's unitary triangularization Theorem [\textit{R. A. Horn, C. R. Jonson}, Matrix analysis. Cambridge etc.: Cambridge University Press (1985; Zbl 0576.15001)] to present their results; for their purposes they use some reduction algorithms on the types of canonical forms given by undirected and directed graphs with no undirected cycles. The paper contains a couple of detailed examples of their applied algorithms of the paper.