Articolo in rivista, 2023, ENG, 10.1007/s10957-022-02153-5

The impact of noise on evaluation complexity: the deterministic trust-region case

Bellavia S.; Gurioli G.; Morini B.; Toint P.L.

Dipartimento di Ingegneria Industriale, Università degli Studi di Firenze, Firenze, Italy; CNR-ISTI, Pisa, Italy; Dipartimento di Ingegneria Industriale, Università degli Studi di Firenze, Firenze, Italy; Namur Center for Complex Systems (naXys), University of Namur, Namur,Belgium

Intrinsic noise in objective function and derivatives evaluations may cause premature termination of optimization algorithms. Evaluation complexity bounds taking this situation into account are presented in the framework of a deterministic trust-region method. The results show that the presence of intrinsic noise may dominate these bounds, in contrast with what is known for methods in which the inexactness in function and derivatives' evaluations is fully controllable. Moreover, the new analysis provides estimates of the optimality level achievable, should noise cause early termination. Numerical experiments are reported that support the theory. The analysis finally sheds some light on the impact of inexact computer arithmetic on evaluation complexity.

Journal of optimization theory and applications

Keywords

Trust-region, Noise, Evaluation complexity, Deterministic

CNR authors

Gurioli Gianmarco

CNR institutes

ISTI – Istituto di scienza e tecnologie dell'informazione "Alessandro Faedo"

ID: 476984

Year: 2023

Type: Articolo in rivista

Creation: 2023-01-26 14:58:23.000

Last update: 2023-02-02 18:15:46.000

External IDs

CNR OAI-PMH: oai:it.cnr:prodotti:476984

DOI: 10.1007/s10957-022-02153-5

Scopus: 2-s2.0-85145939534

ISI Web of Science (WOS): 000911264400002