We show that classical supervised Machine Learning techniques, after trained with a large number of optimal RWA configurations solved via ILP, can rapidly procure the most appropriate RWA configuration to be applied for a new traffic matrix.
Is Machine Learning Suitable for Solving RWA Problems in Optical Networks?
Sebastian Troia;Francesco Musumeci;Guido Maier;
2018-01-01
Abstract
We show that classical supervised Machine Learning techniques, after trained with a large number of optimal RWA configurations solved via ILP, can rapidly procure the most appropriate RWA configuration to be applied for a new traffic matrix.File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
ecoc-paper-2018.pdf
accesso aperto
:
Pre-Print (o Pre-Refereeing)
Dimensione
363.99 kB
Formato
Adobe PDF
|
363.99 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
