Dice Question Streamline Icon: https://streamlinehq.com

Training data scarcity for AI/ML in optical networks

Determine how to obtain and curate training datasets collected from real-world optical network deployments to enable the effective training of AI/ML models for optical network automation, addressing the stated lack of available training data.

Information Square Streamline Icon: https://streamlinehq.com

Background

The paper reviews the state of AI/ML in optical networks and notes persistent gaps that hinder deployment-ready automation. In particular, it cites a lack of real-world datasets as a key barrier to training robust models that can generalize beyond laboratory settings.

Addressing this problem is positioned as foundational for advancing AI-enabled optical network operations and for enabling the active inference research agenda the paper promotes.

References

Further, according to , key AI research challenges remain open: ($i$) Training: Lack of available training datasets from real-world network deployments, ($ii$) Learning: Lack of lifelong (i.e., continual) learning, including AI degradation detection and model adaptation to progressive distribution shift, and ($iii$) Explainability: Lack of trustworthy explainable AI (XAI) due to insufficient transparency of blackbox AI.

— From Artificial Intelligence to Active Inference: The Key to True AI and 6G World Brain [Invited] (2505.10569 - Maier, 29 Apr 2025) in Section 1 (Introduction), Point Alpha — Active Inference in Optical Networks