Machine-Learning-as-a-Service for Optical Networks - Use Cases and Benefits - talk

Aug 3, 2026

Machine Learning (ML) models support the design and operation of optical networks through tasks such as Quality-of-Transmission estimation, device modeling, constellation shaping, and attack or anomaly prediction and detection. Widespread ML use is expected in optical network management, yet model selection, training, evaluation, and deployment often require substantial human effort and empirical decisions. MachineLearning-as-a-Service (MLaaS) reduces this burden by streamlining the creation, assessment, and deployment of ML models for optical networking. The talk covers beneficial use cases, an MLaaS architecture, and performance results for two cases.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages2

Reading Time5 min

AccessFree to download

Overview

Optical-network ML use cases

MLaaS architecture

Performance results

Machine-Learning-as-a-Service for Optical Networks: Use Cases and Benefits Downloaded from: [https://research.chalmers.se](https://research.chalmers.se), 2023-10-28 13:59 UTC Citation for the original published paper (version of record): Natalino Da Silva, C., Mohammadiha, N., Panahi, A. et al (2023). Machine-Learning-as-a-Service for Optical Networks: Use Cases and Benefits. Proceedings of the 23rd International Conference on Transparent Optical Networks, 2023 N. B. When citing this work, cite the original published paper. research.chalmers.se offers the possibility of retrieving research publications produced at Chalmers University of Technology. It covers all kind of research output: articles, dissertations, conference papers, reports etc. since 2004. research.chalmers.se is administrated and maintained by Chalmers Library (article starts on next page) Machine-Learning-as-a-Service for Optical Networks: Use Cases and Benefits C. Natalino1, N. Mohammadiha2,3, A. Panahi2, and P. Monti1 1Department of Electrical Engineering, Chalmers University of Technology, Gothenburg, Sweden 2Department of Computer Science and Engineering, Chalmers University of Technology, Gothenburg, Sweden 3Ericsson Research, Gothenburg, Sweden ABSTRACT Machine Learning (ML) models have been a valuable tool to assist on the design and operation of optical networks. Several use cases have benefited from ML models, such as Quality-of-Transmission (QoT) estimation, device modeling, constellation shaping, and attack/anomaly prediction/detection. ML models are expected to be ubiquitous in optical network management and operations thereof. However, the amount of human intervention and empirical decisions needed to select the exact ML model, train and evaluate its performance, and ultimately deploy and use the model, may become a bottleneck for widespread ML use in optical networks. MachineLearning-as-a-Service (MLaaS) has the potential to greatly reduce human intervention and empirical decisions during the creation, evaluation, and deployment of ML models. In this talk, we will firstly discuss optical network use cases that can benefit from MLaaS. Then, we detail our proposed architecture for MLaaS. Finally, performance results for two use cases will be presented.

FAQs

Which optical network tasks can benefit from MLaaS-supported machine learning models?

Tasks include Quality-of-Transmission (QoT) estimation, device modeling, constellation shaping, and attack/anomaly prediction or detection.

Why can human effort limit widespread ML adoption in optical networks?

Selecting the right ML model, training and evaluating performance, and deploying it can require significant human intervention and empirical decisions.

What does the MLaaS approach aim to improve in the ML lifecycle?

MLaaS reduces the need for manual steps during creation, evaluation, and deployment of ML models, helping integrate them more broadly into optical network operations.

Aug 3, 2026

Upload by Olivia Brown