scientific article; zbMATH DE number 3882209
[For the entire collection see Zbl 0541.00004.] Conjugate gradient algorithms, using O(n) operations/iterations, and variable metric algorithms, using \(O(n^ 2)\) operations/iterations, minimize quadratic objective functions by minimizing sequentially along lines in conjugate directions. While the concept of conjugacy used in these algorithms is invariant under affine transformations \(S(w)=x_ 0+Jw\), it is a special case of an older concept from projective geometry which is invariant under the larger group of collineations \(S(w)=x_ 0+Jw/(1+h^ Tw)\). This paper suggests algorithms using this more general concept of conjugacy.
- Some investigations in a new algorithm for nonlinear optimization based on conic models of the objective function
- Generalized conjugate directions
- On accelerating Newton's method based on a conic model
- A new conic method for unconstrained minimization
- Conjugate direction methods and polarity for quadratic hypersurfaces
- A new reprojection of the conjugate directions
- scientific article; zbMATH DE number 992794 (Why is no real title available?)
- Relación entre conos de direcciones decrecientes y conos de direcciones de descenso
- scientific article; zbMATH DE number 3932292 (Why is no real title available?)
- scientific article; zbMATH DE number 3963115 (Why is no real title available?)
- Algorithms with Conic Termination for Nonlinear Optimization
- Conjugated of conics
- Conjugate direction methods in non-Hilbert spaces
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3216440)