Networks are the enabler of the Information Age: They have been powering a large portion of the technological advancements of the past decade and they have become national critical infrastructure and very sophisticated. They are intrinsically distributed, heterogeneous, and dynamic, thus making their design, management and troubleshooting very challenging. AI-ML offer a series of powerful tools and methods to tackle these challenges, ranging from network assurance, security, intelligent control to self-healing and from anomaly detection, root cause detection, prediction to forecasting. However, a large portion of this field remains to be explored.
In this track we would like to explore the following topics:
- Overview of landscape in terms of interoperability, integration, automation of AI in operations, multi-vendor, semantics...
- The duality of "AI to solve networking problems and networking to solve AI problems"
- Reality checks on the deployment and level of development, adoption, and integration of AI in the real world
- Clarifying current limits and challenges in AI for Networks
- Emerging new approaches (e.g. new forms and combinations developed for or customized for networks)