In this paper we study new preconditioners to be used within the Nonlinear Conjugate Gradient (NCG) method, for large scale unconstrained optimization. The rationale behind our proposal draws inspiration from quasi– Newton updates, and its aim is to possibly approximate in some sense the inverse of the Hessian matrix. In particular, at the current iteration of the NCG we consider some preconditioners based on new low–rank quasi–Newton symmetric updating formulae, obtained as by–product of the NCG method at the previous steps. The results of an extensive numerical experience are also reported, showing the effectiveness, the efficiency and the robustness of this approach, which suggests promising guidelines for further studies.

Novel Quasi–Newton updates for Preconditioned Nonlinear Conjugate Gradient methods

FASANO, Giovanni;
2017-01-01

Abstract

In this paper we study new preconditioners to be used within the Nonlinear Conjugate Gradient (NCG) method, for large scale unconstrained optimization. The rationale behind our proposal draws inspiration from quasi– Newton updates, and its aim is to possibly approximate in some sense the inverse of the Hessian matrix. In particular, at the current iteration of the NCG we consider some preconditioners based on new low–rank quasi–Newton symmetric updating formulae, obtained as by–product of the NCG method at the previous steps. The results of an extensive numerical experience are also reported, showing the effectiveness, the efficiency and the robustness of this approach, which suggests promising guidelines for further studies.
2017
11
File in questo prodotto:
File Dimensione Formato  
11590_2016_1060_Author.pdf

accesso aperto

Descrizione: Eproofs della rivista inviate agli autori dopo l'accettazione del paper
Tipologia: Documento in Pre-print
Licenza: Licenza non definita
Dimensione 696.49 kB
Formato Adobe PDF
696.49 kB Adobe PDF Visualizza/Apri

I documenti in ARCA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10278/3677533
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 18
  • ???jsp.display-item.citation.isi??? 14
social impact