Edge AI: Leveraging the Full Potential of Deep Learning

Edge AI: A survey

“…The algorithms that make up AI are finding their way into a growing number of the goods and services that we use. Edge AI refers to AI that is implemented directly into a smart device [148]. The way in which we make use of it and the technical potential presented by the study raises a number of issues to understand the problems, the technologies that are utilized, the use cases, the stakeholders, and the chances for the future of computing [149].…”

“…Computing Resources: Edge computing devices, such as IoT sensors and edge gateways, often have fewer processing resources than central clouds [148]. As a result, it might be difficult to run sophisticated AI models and techniques on such devices because they could demand more resources than are really accessible [149].…”

“…AI systems are very helpful in settings with real challenges for end customers because they can interpret analogue types of metadata such as language, images, audio, scents, temperatures, emotions, and more [148]. Due to delay, speed, and confidentiality issues, it would be difficult, if not inconceivable, to install these AI applications in a centralised cloud or business data centre [149].…”