The selection of cuts to be added to the current LP relaxation is one of the most critical task in Branch-and-Cut methods, since it strongly affects the performances of the algorithm. Recently, machine learning techniques have become popular to define effective cut selection strategies. In this paper we explore the possibility of selecting cuts by ranking them via support vector regression.
Machine learning techniques for Branch-and-Cut methods: the selection of cutting planes
Marcello SammarraMembro del Collaboration Group
;
2023
Abstract
The selection of cuts to be added to the current LP relaxation is one of the most critical task in Branch-and-Cut methods, since it strongly affects the performances of the algorithm. Recently, machine learning techniques have become popular to define effective cut selection strategies. In this paper we explore the possibility of selecting cuts by ranking them via support vector regression.File in questo prodotto:
File | Dimensione | Formato | |
---|---|---|---|
PaperNumtaGMS23-R1.pdf
solo utenti autorizzati
Tipologia:
Documento in Pre-print
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
311.83 kB
Formato
Adobe PDF
|
311.83 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.