Preconditioned conjugate gradient method
ID: preconditioned-conjugate-gradient-method
For a real symmetric positive-definite matrix , choose an invertible change of variables and solveThe transformed matrix is symmetric positive definite because for nonzero . The aim is to improve its condition number or eigenvalue clustering while keeping the transformation inexpensive. The change-of-variables matrix itself need not be symmetric.
A common convention starts instead from a symmetric positive-definite approximate matrix and uses conceptually. If is a Cholesky decomposition, choosing gives an equivalent symmetric transformation. Practical iterations can apply solves with or its factors without explicitly constructing a matrix square root. Applying ordinary Euclidean conjugate gradients directly to the generally nonsymmetric matrix would require accounting for the appropriate transformed inner product.
New to topics? Read the docs here!