BandPC: Learning Residual-Band Preconditioner Combinations for Flexible Conjugate Gradient Solvers (Q7360725)
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:
No description defined
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | BandPC: Learning Residual-Band Preconditioner Combinations for Flexible Conjugate Gradient Solvers |
No description defined |
Statements
BandPC: Learning Residual-Band Preconditioner Combinations for Flexible Conjugate Gradient Solvers (English)
0 references
29 September 2026
0 references
math.NA
0 references
D. M. Li
0 references
Shang-Tian Yang
0 references
Xin Qiu
0 references
We develop BandPC, a graph neural network-based framework to predict optimal combinations of classical preconditioners for the flexible conjugate gradient method in sparse symmetric positive definite linear systems. Our approach exploits FCG's ability to switch preconditioners across iterations and learns from matrix structure to find a promising sequence.
0 references
Preconditioning
0 references
Flexible Conjugate Gradient
0 references
Residual Band
0 references
1 reference
1 reference
1 reference
1 reference
1 reference
1 reference
1 reference
1 reference
1 reference
1 reference
