Parallel variable metric algorithms for unconstrained optimization (Q3738945)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 3972656
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Parallel variable metric algorithms for unconstrained optimization |
scientific article; zbMATH DE number 3972656 |
Statements
Parallel variable metric algorithms for unconstrained optimization (English)
0 references
1985
0 references
unconstrained optimization
0 references
parallel computing
0 references
parallel variable metric algorithm
0 references
0 references
0 references
0.8581045866012573
0 references
0.8548982739448547
0 references
0.850944459438324
0 references
0.8488924503326416
0 references
0.8467135429382324
0 references
The purpose of this paper is to present algorithms for unconstrained optimization which could be used to exploit effectively the parallel computing capabilities. The author thoroughly analyzed Straeter's ideas of parallel variable metric algorithm for unconstrained optimization. He then applies them to Huang's class of updating formulas and to Broyden's 1965 method. He shows that under certain assumptions Straeter's algorithm is the only parallel extension within Huang's class of updating formulas with the property of quadratic termination in one major iteration. The parallel algorithms were tested on four well-known sample functions. The numerical experiments indicated that the parallel variable metric algorithms exhibit considerably better convergence than the (sequential) variable metric algorithm, i.e., BFGS algorithm.
0 references