Communication network

Machine learning systems can benefit from transfer learning of knowledge in training new models. Cognitive Autonomous Network OAM systems may utilise machine learning (ML) elements such as neural networks. MLTL producers create/possess knowledge (e.g. statistics or pre-trained models) and they may p...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Stephen Mwanje, Abdelrahman Abdelkader
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Machine learning systems can benefit from transfer learning of knowledge in training new models. Cognitive Autonomous Network OAM systems may utilise machine learning (ML) elements such as neural networks. MLTL producers create/possess knowledge (e.g. statistics or pre-trained models) and they may publish (0) the knowledge or information about it (e.g. task, domain, context descriptions) to repositories. MLTL consumers and/or recipients can discover (1) information about knowledge by requesting reports from producers and/or repositories. The MLTL consumers/recipients may also request all/part of the actual knowledge (2) in order to use it in transfer learning (TL).