WS-05: DISTRIBUTED AI/ML AT THE RESOURCE CONSTRAINED EDGE

The feasibility and scalability of Artificial Intelligence and Machine Learning (AI/ML) algorithms at the edge networks faces three principal challenges: (i) communication and computation cost, (ii) heterogeneity and time-varying nature of edge devices, and (iii) heightened security and privacy concerns given the vulnerability of edge devices. In this context, research is needed to tailor AI/ML mechanisms to edge computing to harvest heterogeneous resources including computing power, storage, battery, networking resources, etc., scattered across end devices, edge servers, and cloud with
minimum amount of communication cost. Despite the rapidly growing scientific body of work to implement AI/ML mechanisms at the edge networks, the area is still nascent, and is conducted by many researchers whose instincts come from classical applications such as communications and networking.  By exposing the challenges and ideas of modern AI/ML systems to edge computing, this workshop aims to provide a unique opportunity to shape the growth of the field and maximize the chances of its impact to meet AI/ML with edge computing. This workshop will explore the AI/ML mechanism over edge networks by exploiting techniques including communication efficient distributed AI/ML, conditional computation, and decentralized learning.