  


Linearizing the Method of Conjugate Gradients
Serge Gratton(grattoncerfacs.fr) Abstract: The method of conjugate gradients (CG) is widely used for the iterative solution of large sparse systems of equations $Ax=b$, where $A\in\Re^{n\times n}$ is symmetric positive definite. Let $x_k$ denote the $k$th iterate of CG. In this paper we obtain an expression for $J_k$, the Jacobian matrix of $x_k$ with respect to $b$. We use this expression to obtain computable bounds on the spectral norm condition number of $x_k$, and to design algorithms to compute or estimate $J_kv$ and $J_k^Tv$ for a given vector $v$. We also discuss several applications in which these ideas may be used. Numerical experiments are performed to illustrate the theory. Keywords: Conjugate Gradients Algorithm, Lanczos Algorithm, Perturbation Analysis, Linearization, Automatic Differentiation Category 1: Applications  Science and Engineering (Basic Sciences Applications ) Category 2: Nonlinear Optimization (Unconstrained Optimization ) Citation: Download: [PDF] Entry Submitted: 09/09/2012 Modify/Update this entry  
Visitors  Authors  More about us  Links  
Subscribe, Unsubscribe Digest Archive Search, Browse the Repository

Submit Update Policies 
Coordinator's Board Classification Scheme Credits Give us feedback 
Optimization Journals, Sites, Societies  