Adaptive cubic regularization methods for solving nonconvex problems need the efficient computation of the trial step, involving the minimization of a cubic model. We propose a new approach in which this model is minimized in a low dimensional subspace that, in contrast to classic approaches, is reused for a number of iterations. Whenever the trial step produced by the low-dimensional minimization process is unsatisfactory, we employ a regularized Newton step whose regularization parameter is a by-product of the model minimization over the low-dimensional subspace. We show that the worst-case complexity of classic cubic regularized methods is preserved, despite the possible regularized Newton steps. We focus on the large class of problems for which (sparse) direct linear system solvers are available and provide several experimental results showing the very large gains of our new approach when compared to standard implementations of adaptive cubic regularization methods based on direct linear solvers. Our first choice as projection space for the low-dimensional model minimization is the polynomial Krylov subspace; nonetheless, we also explore the use of rational Krylov subspaces in case where the polynomial ones lead to less competitive numerical results.

Regularized methods via cubic model subspace minimization for nonconvex optimization

Porcelli M.;Simoncini V.
2025

Abstract

Adaptive cubic regularization methods for solving nonconvex problems need the efficient computation of the trial step, involving the minimization of a cubic model. We propose a new approach in which this model is minimized in a low dimensional subspace that, in contrast to classic approaches, is reused for a number of iterations. Whenever the trial step produced by the low-dimensional minimization process is unsatisfactory, we employ a regularized Newton step whose regularization parameter is a by-product of the model minimization over the low-dimensional subspace. We show that the worst-case complexity of classic cubic regularized methods is preserved, despite the possible regularized Newton steps. We focus on the large class of problems for which (sparse) direct linear system solvers are available and provide several experimental results showing the very large gains of our new approach when compared to standard implementations of adaptive cubic regularization methods based on direct linear solvers. Our first choice as projection space for the low-dimensional model minimization is the polynomial Krylov subspace; nonetheless, we also explore the use of rational Krylov subspaces in case where the polynomial ones lead to less competitive numerical results.
2025
Istituto di Scienza e Tecnologie dell'Informazione "Alessandro Faedo" - ISTI
Adaptive regularization methods, Nonconvex optimization, Secant methods, Krylov subspaces, Worst-case iteration complexity
File in questo prodotto:
File Dimensione Formato  
Porcelli et al_Springer 2025.pdf

accesso aperto

Descrizione: Regularized methods via cubic model subspace minimization for nonconvex optimization
Tipologia: Versione Editoriale (PDF)
Licenza: Creative commons
Dimensione 753.07 kB
Formato Adobe PDF
753.07 kB Adobe PDF Visualizza/Apri
Porcelli et al_Springer 2025_Correction.pdf

accesso aperto

Descrizione: Correction: Regularized methods via cubic model subspace minimization for nonconvex optimization
Tipologia: Versione Editoriale (PDF)
Licenza: Creative commons
Dimensione 440.69 kB
Formato Adobe PDF
440.69 kB Adobe PDF Visualizza/Apri

I documenti in IRIS 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/20.500.14243/542901
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 0
social impact