Show simple item record

Modified Cholesky Algorithms: A Catalog with New Approaches

dc.contributor.authorFang, Haw-ren
dc.contributor.authorO'Leary, Dianne P.
dc.date.accessioned2006-08-08T17:32:51Z
dc.date.available2006-08-08T17:32:51Z
dc.date.issued2006-08-08
dc.identifier.urihttp://hdl.handle.net/1903/3674
dc.description.abstractGiven an $n \times n$ symmetric possibly indefinite matrix $A$, a modified Cholesky algorithm computes a factorization of the positive definite matrix $A+E$, where $E$ is a correction matrix. Since the factorization is often used to compute a Newton-like downhill search direction for an optimization problem, the goals are to compute the modification without much additional cost and to keep $A+E$ well-conditioned and close to $A$. Gill, Murray and Wright introduced a stable algorithm, with a bound of $\|E\|_2=O(n2)$. An algorithm of Schnabel and Eskow further guarantees $\|E\|_2=O(n)$. We present variants that also ensure $\|E\|_2=O(n)$. Mor\'{e} and Sorensen and Cheng and Higham used the block $LBL^T$ factorization with blocks of order $1$ or $2$. Algorithms in this class have a worst-case cost $O(n3)$ higher than the standard Cholesky factorization, We present a new approach using an $LTL^T$ factorization, with $T$ tridiagonal, that guarantees a modification cost of at most $O(n2)$.en
dc.format.extent433166 bytes
dc.format.mimetypeapplication/pdf
dc.language.isoen_USen
dc.relation.ispartofseriesUM Computer Science Departmenten
dc.relation.ispartofseriesCS-TR-4807en
dc.relation.ispartofseriesUMIACSen
dc.relation.ispartofseriesUMIACS-TR-2006-27en
dc.titleModified Cholesky Algorithms: A Catalog with New Approachesen
dc.typeTechnical Reporten


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record