“…To run neural networks, recent developments in massively parallel GPUs have been applied. The proliferation of IoT-connected devices is largely responsible for the current era's unprecedented growth in data volume [34]. Now that we have the data and the devices required to implement AI models at the edge, we can start doing so in practically every facet of the industry, thanks to the proliferation of sensors, smart cameras, robotics, and other data-gathering tools [35].…”