Demystifying the Edge AI Paradigm

The device ecosystem is growing rapidly these days. We are being bombarded with scores of purpose-agnostic and specific devices. There are wearable, wireless, mobile, fixed, portable, implantable, and nomadic devices serving humanity in our everyday decisions, deals, and deeds. More resources are accommodated in personal and professional devices and hence these empowered devices are capable of doing data captured, storage, processing, and analytics. Edge computing aspires in the coordination and collaboration of edge devices for processing the data generated in proximity; wherein artificial intelligence aims for creating intelligence on these devices by learning from those data. Edge computing is the computing capability and facility being provided by edge/fog devices. Edge sensors and devices are projected to be in the billions in the years to come; the amount of data getting generated by edge devices is humungous as there is a realization that every type of edge data comprises useful and usable knowledge. Cloud computing is quite centralized, whereas edge or fog computing is a distributed model. Multiple heterogeneous devices cooperate with one another to do edge data collection, storage, and processing. Collection of real-time data, storage, processing, analytics, making the actionable result, and action are being made possible with the realization of edge cloud infrastructures.