Iterative Methods

The conjugate gradient algorithm can be interpreted as the following optimization problem: minimize defined by

     

where and are symmetric and positive definite.

At each iteration is minimized along an -conjugate direction, constructing orthogonal residuals:

     

where is a Krylov subspace:

     

Minimum residual algorithms work by minimizing the Euclidean norm over . At each iteration, is the vector in that gives the smallest residual.

The biconjugate gradient algorithm belongs to a more general class of Petrov-Galerkin methods, where orthogonality is enforced in a different -dimensional subspace ( remains in ):