Learning-enabled Networked Systems

In the age of ubiquitous connectivity, networked systems, from edge computing platforms and wireless networks to expansive social networks, have grown in complexity and scale, impacting various sectors of society and industry. Traditional network management and optimization techniques, often reliant on fixed parameters and static models, struggle to cope with the dynamic and multifaceted nature of modern networked environments. Machine learning, with its ability to adaptively process vast amounts of data and refine its models over time, emerges as a game-changer, o↵ering the potential to transform these systems into adaptive, ecient, and more resilient entities. Integrating machine learning into networked systems can enable them to selfoptimize, predict potential failures, and even facilitate usercentric adaptations, paving the way for a new generation of intelligent and responsive networks.