1977
DOI: 10.1137/0714072
|View full text |Cite
|
Sign up to set email alerts
|

Safe Starting Regions for Iterative Methods

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
37
0

Year Published

1982
1982
2006
2006

Publication Types

Select...
8
1

Relationship

0
9

Authors

Journals

citations
Cited by 103 publications
(39 citation statements)
references
References 7 publications
0
37
0
Order By: Relevance
“…These methods find all solutions of (1) in a finite domain. As discussed by Moore and Jones ( [9], p. 1056), three possibilities arise when examining such a finite domain. One is sometimes able to show that no solution exists, or that at least one solution exists.…”
Section: In a Family Of Systems H(xt) A Frequently Used Embedding Imentioning
confidence: 99%
“…These methods find all solutions of (1) in a finite domain. As discussed by Moore and Jones ( [9], p. 1056), three possibilities arise when examining such a finite domain. One is sometimes able to show that no solution exists, or that at least one solution exists.…”
Section: In a Family Of Systems H(xt) A Frequently Used Embedding Imentioning
confidence: 99%
“…In this case, however, the interval iteration process (4.1) is 6inite, and will tde'nunate in either the condition ztaled. The difficulty concerning existence can be resolved by applying some other existence test than (4.4) to X = XN, a region which will be smaller than X 0 unless the iteration stalled, or x can be divided into subregions for further investigation, using the algorithm of Moore and Jones, which has also been shown to be finite if G is sufficiently fine [13], [14].…”
Section: -3-' IImentioning
confidence: 99%
“…(Y°m ay be the limit Y* if obtained in a finite number of steps, or otherwise.) Possible alternatives in this situation include partition of the resulting interval Y°i nto subintervals for further examination, a strategy developed by Moore and Jones [9], or acceptance of Y° as a generalized or pseudosolution (relative to the initial interval Y0) of the fixed point problem. This latter choice may be useful in the development of an interval version of regularization of solutions of ill-posed problems.…”
mentioning
confidence: 99%