Artificial Intelligence at the Edge in a 5G world - Seguridad Mania.com - España y América Latina

Deploying AI/ML solutions in latency-sensitive use cases requires a new solution architecture approach for many businesses. Fast computational units (i.e. GPUs) and low-latency connections (i.e. 5G) allow for AI/ML models to be executed outside the sensors/actuators (e.g. cameras & robotic arms). This reduces costs through lower hardware complexity as well as compute resource sharing amongst the IoT fleet. Strict AI responsiveness requirements that before required IoT AI model embedding can now be met with co-located GPUs (e.g. on the same factory building) as the sensors and actuators. An example of this is the robot ‘dummification’ trend that is currently being observed for factory robotics with a view to reducing robot unit costs and fleet management. In this webinar we will explore some real-life scenarios in which GPUs and low-latency connectivity can unlock previously prohibitively expensive solutions now available for businesses to put in place and lead the 4th industrial revolution.