AIatEDGE

Website:               https://aiatedge.eu/           Social media:                                Contact:    AIatEDGE-Contact@5g-ppp.eu

Abstract:

Artificial Intelligence has become a major innovative force and it is one of the pillars of the fourth industrial revolution. This trend has been acknowledged also by the European Commission that has already pointed out how high-performance, intelligent, and secure networks are fundamental for the development and evolution of the multi-service Next Generation Internet (NGI). While great progress has been done during the last years with respect to the accuracy and performance of AI-enabled platforms, their integration in potentially autonomous decision-making systems or even critical infrastructures requires end-to-end quality assurance.

AI@EDGE addresses the challenges harnessing the concept of “reusable, secure, and trustworthy AI for network automation”. In AI@EDGE European industries, academics and innovative SMEs commit to achieve an EU-wide impact on industry-relevant aspects of the AI-for-networks and networks-for-AI paradigms in beyond 5G systems. Cooperative perception for vehicular networks, secure, multi-stakeholder AI for IIoT, aerial infrastructure inspections, and in-flight entertainment are the uses cases targeted by AI@EDGE to maximise the commercial, societal, and environmental impact. To achieve the goal, AI@EDGE targets significant breakthroughs in two fields: (i) general-purpose frameworks for closed-loop network automation capable of supporting flexible and programmable pipelines for the creation, utilization, and adaptation of the secure, reusable, and trustworthy AI/ML models; and (ii) converged connect-compute platform for creating and managing resilient, elastic, and secure end-to-end slices capable of supporting a diverse range of AI-enabled network applications.

Main Objectives:

AI@EDGE will build a platform and tools enabling secure and automated roll-out of large-scale edge and cloud compute infrastructures, with close to zero-touch of the underlying heterogeneous MEC resources (network, storage, and compute resources). The serverless computing paradigm will be used to automate the deployment of distributed applications at scale. This will be combined with native hardware acceleration (e.g. GPU and FPGA) and cross-layer, multi-connectivity radio access. To do this, we will develop a systems architecture and the necessary methods to automatically manage heterogeneous MEC resources. The project includes the development of low-overhead and scalable communication schemes suitable for running distributed algorithms at a massive scale under highly varying networking conditions. Moreover, we will address fundamentally important security aspects, covering both service isolation and protection required for secure federated learning and data sharing, as well as security for AI/ML models that may be vulnerable to manipulation attacks.

List of Participants