An adaptive extraction of generalized eigensubspace by using exact nested orthogonal complement structure (Q2393098)

The authors propose a scheme for generalized minor eigensubspace extraction. The proposed scheme is based on the dimension reduction technique, and allows to employ any algorithm for estimating the first minor generalized eigenvector. The authors give a pair of such algorithms and analyze their convergence properties in the case where the generalized eigenvalues are not necessarily distinct. Finally, they present an implementation of the proposed scheme for estimating an orthonormal basis of the generalized minor subspace. The eigensubspace extraction is illustrated on numerical examples.