We propose an evolutionary transfer learning approach for QoT estimation in multi-domain optical networks. The results demonstrate that our approach can reduce the amounts of required training data by 10× while achieving accuracies of > 90%.

© 2020 The Author(s)

PDF Article
More Like This
Collaborative Learning in Multi-Domain Optical Networks

Xiaoliang Chen, Roberto Proietti, Che-Yu Liu, and S. J. Ben Yoo
T3C.2 Asia Communications and Photonics Conference (ACPC) 2020

Machine-Learning-Aided Service Provisioning in Multi-Domain Optical Networks

Roberto Proietti, Xiaoliang Chen, Gengchen Liu, Hongbo Lu, and S.J. Ben Yoo
NeT1D.4 Photonic Networks and Devices (Networks) 2019

Transfer learning aided concurrent multi-alarm prediction in optical transport networks

Bing Zhang, Yongli Zhao, Yajie Li, and Jie Zhang
M4A.197 Asia Communications and Photonics Conference (ACPC) 2020


You do not have subscription access to this journal. Citation lists with outbound citation links are available to subscribers only. You may subscribe either as an OSA member, or as an authorized user of your institution.

Contact your librarian or system administrator
Login to access OSA Member Subscription