Minimum relative error approximations for 1/t (Q1111315): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
ReferenceBot (talk | contribs) Changed an Item |
||
(3 intermediate revisions by 2 users not shown) | |||
Property / reviewed by | |||
Property / reviewed by: Q688742 / rank | |||
Property / reviewed by | |||
Property / reviewed by: Dietrich Braess / rank | |||
Normal rank | |||
Property / MaRDI profile type | |||
Property / MaRDI profile type: MaRDI publication profile / rank | |||
Normal rank | |||
Property / cites work | |||
Property / cites work: Q5543516 / rank | |||
Normal rank |
Latest revision as of 10:57, 19 June 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Minimum relative error approximations for 1/t |
scientific article |
Statements
Minimum relative error approximations for 1/t (English)
0 references
1988
0 references
There are papers which provoke many letters to the editors of a journal. This paper is a good candidate. Chebyshev and Bernstein provided formulas for the best uniform approximation of 1/t by polynomials. The analysis of the best relative approximation for a real interval is even simpler than that of the best absolute error. The results of the well known theory are the basis of the concept of semiiterative methods and are also crucial for understanding the good performance of conjugate gradient methods. Therefore, in that framework there are also many investigations of the approximation in more general regions in the complex plane.
0 references
minimum relative error approximations
0 references
best relative approximation
0 references
best absolute error
0 references
semiiterative methods
0 references
conjugate gradient methods
0 references