Polynomial programming using multi-homogeneous polynomial continuation (Q1195745): Difference between revisions
From MaRDI portal
Created a new Item |
Added link to MaRDI item. |
||
links / mardi / name | links / mardi / name | ||
Revision as of 05:41, 31 January 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Polynomial programming using multi-homogeneous polynomial continuation |
scientific article |
Statements
Polynomial programming using multi-homogeneous polynomial continuation (English)
0 references
18 January 1993
0 references
In this note a polynomial programming problem of the type \(\min_ x\{\theta(x)\mid\) \(h(x)=0\}\), where \(\theta\), \(h\) are polynomial functions \(\theta: E^ n\to E\), \(h: E^ n\to E^ q\), is considered. Traditional constructions of the polynomial continuation approach need the computation of the complete list of geometrically isolated solutions of the first order optimality conditions (1) \(\nabla\theta(x)+r^ t\nabla h(x)=0\), \(h(x)=0\). In this note the variables \(x\) are partitioned into \(m\) nonempty sets and new variables are created in such a way that the new system generated from (1) will be homogeneous. This homogenization gives a possibility to reduce the number of paths needed to solve the problem.
0 references
global optimization
0 references
globally convergent
0 references
homotopy algorithm
0 references
\(m\)- homogeneous
0 references
multi-homogeneous
0 references
parallel optimization
0 references
polynomial continuation
0 references
polynomial program
0 references
polynomial programming
0 references
isolated solutions
0 references
homogenization
0 references