The Influence of the Cluster Point on Rounding Errors and Sensitivity in the Spectral Limited-Memory Preconditioner
The spectral limited-memory preconditioner (sLMP) clusters leading eigenvalues of symmetric positive definite matrices to accelerate conjugate gradient (CG) convergence. In practice, the cluster point is often chosen to be unity. In some cases, however, this choice can fail to accelerate convergence relative to unpreco...