Self-Normalizing Neural Network, Enabling One Shot Transfer Learning for Modeling EDFA Wavelength Dependent Gain

2023/08/04 by Raj, Agastya, Wang, Zehao, Slyne, Frank +3
#FOS: Computer and information sciences #Machine Learning (cs.LG) #Networking and Internet Architecture (cs.NI)

paper · doi:10.48550/arxiv.2308.02233

Abstract

We present a novel ML framework for modeling the wavelength-dependent gain of multiple EDFAs, based on semi-supervised, self-normalizing neural networks, enabling one-shot transfer learning. Our experiments on 22 EDFAs in Open Ireland and COSMOS testbeds show high-accuracy transfer-learning even when operated across different amplifier types.

Related

• Transfer Learning for EDFA Gain Modeling: A Semi-Supervised Approach Using Internal Amplifier Features
• Generalized few-shot transfer learning architecture for modeling the EDFA gain spectrum
• Machine learning-based EDFA Gain Model Generalizable to Multiple\n Physical Devices
• Physics-Informed Machine Learning for EDFA: Parameter Identification and Gain Estimation
• Transfer Learning for Transformer-Based Modeling of Nonlinear Pulse Evolution in Er-Doped Fiber Amplifiers