Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems

02/25/2022

∙

We introduce the domain adaptation and randomization approach for calibrating neural network-based equalizers for real transmissions, using synthetic data. The approach renders up to 99% training process reduction, which we demonstrate in three experimental setups.

READ FULL TEXT

Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems

Unsupervised Domain Adaptation with Deep Neural-Network

PieTrack: An MOT solution based on synthetic data training and self-supervised domain adaptation

DeceptionNet: Network-Driven Domain Randomization

A Novel Domain Adaptation Framework for Medical Image Segmentation

Neural network based limiter with transfer learning

VoloGAN: Adversarial Domain Adaptation for Synthetic Depth Data

VADRA: Visual Adversarial Domain Randomization and Augmentation

Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems

Related Research

Unsupervised Domain Adaptation with Deep Neural-Network

PieTrack: An MOT solution based on synthetic data training and self-supervised domain adaptation

DeceptionNet: Network-Driven Domain Randomization

A Novel Domain Adaptation Framework for Medical Image Segmentation

Neural network based limiter with transfer learning

VoloGAN: Adversarial Domain Adaptation for Synthetic Depth Data

VADRA: Visual Adversarial Domain Randomization and Augmentation