Add to BibSonomyAbstract
A Bayesian neural network-based uncertainty analysis technique is proposed for failure prediction in optical transport network, which can not only achieve F1-score up to 99.5%, but also give an uncertainty quantification for the prediction results.
© 2021 The Author(s)
PDF ArticleMore Like This
Machine Learning for Optical Layer Failure Management
Danshi Wang, Dongdong Wang, Chunyu Zhang, Lingling Wang, Songlin Liu, and Min Zhang
T3A.3 Optoelectronics and Communications Conference (OECC) 2021
Mastering Parameter Uncertainty in Monitoring-Enabled Optical Networks using Bayesian Inference
Antonio Massaro, Matteo Lonardi, and Petros Ramantanis
W3C.4 Asia Communications and Photonics Conference (ACP) 2021
Semi-supervised Anomaly Detection with Imbalanced Data for Failure Detection in Optical Networks
SongLin Liu, Danshi Wang, Chunyu Zhang, Lingling Wang, and Min Zhang
Th1A.24 Optical Fiber Communication Conference (OFC) 2021