Convergence of Descent Optimization Algorithms under Polyak-Łojasiewicz-Kurdyka Conditions
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909488802430976 |
|---|---|
| author | Bento, G. C. Mordukhovich, B. S. Mota, T. S. Nesterov, Yu. |
| author_facet | Bento, G. C. Mordukhovich, B. S. Mota, T. S. Nesterov, Yu. |
| contents | This paper develops a comprehensive convergence analysis for generic classes of descent algorithms in nonsmooth and nonconvex optimization under several conditions of the Polyak-Łojasiewicz-Kurdyka (PLK) type. Along other results, we prove the finite termination of generic algorithms under the PLK conditions with lower exponents. Specifications are given to establish new convergence rates for inexact reduced gradient methods and some versions of the boosted algorithm in DC programming. It is revealed, e.g., that the lower exponent PLK conditions for a broad class of difference programs are incompatible with the gradient Lipschitz continuity for the plus function around a local minimizer. On the other hand, we show that the above inconsistency observation may fail if the Lipschitz continuity is replaced by merely the gradient continuity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_00812 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Convergence of Descent Optimization Algorithms under Polyak-Łojasiewicz-Kurdyka Conditions Bento, G. C. Mordukhovich, B. S. Mota, T. S. Nesterov, Yu. Optimization and Control 65K05, 65K10, 90C26, 47N10 This paper develops a comprehensive convergence analysis for generic classes of descent algorithms in nonsmooth and nonconvex optimization under several conditions of the Polyak-Łojasiewicz-Kurdyka (PLK) type. Along other results, we prove the finite termination of generic algorithms under the PLK conditions with lower exponents. Specifications are given to establish new convergence rates for inexact reduced gradient methods and some versions of the boosted algorithm in DC programming. It is revealed, e.g., that the lower exponent PLK conditions for a broad class of difference programs are incompatible with the gradient Lipschitz continuity for the plus function around a local minimizer. On the other hand, we show that the above inconsistency observation may fail if the Lipschitz continuity is replaced by merely the gradient continuity. |
| title | Convergence of Descent Optimization Algorithms under Polyak-Łojasiewicz-Kurdyka Conditions |
| topic | Optimization and Control 65K05, 65K10, 90C26, 47N10 |
| url | https://arxiv.org/abs/2407.00812 |