Issue link: https://resources.mouser.com/i/1481902
Intelligence at the Edge 16 Discussing the Frontier Edges of AI and ML Paul Golata C ollege professors are entrusted by society and institutions with shaping young minds. Their job is to help them learn and get them excited about expanding their knowledge on new subjects, to pour into them from a deep reservoir of knowledge on various subject matters and test them to demonstrate that they grasp the material. Outside the classroom, another type of learning brings us to the edge of new frontiers. Machine learning (ML), a subset of artificial intelligence (AI), employs algorithms to extract information from raw data and represent it in some type of model. It helps a machine to learn without additional direct human programming. This article is a technical perspective and discussion of the current state of AI and ML, articulating how Edge AI is penetrating the markets of today and tomorrow, and how it will positively impact our lives in the coming years. A Current Look at ML The starting point is raw data. Edge computing brings computational power and performs it where data is initiated and collected. The benefit to this is that it gives users the ability to access and utilize the data in real-time. Coupled with the power of AI and ML, it allows designers to unlock the next wave of intelligent devices, the so-called Artificial Internet of Things (AIoT). Locally sampling customized, scalable, secure, and real-time sensor data facilitates fast decision-making and feedback. The ML algorithm continually optimizes itself through recursive learning. The goal is to have enough computing resources to perform Edge AI while simultaneously optimizing hardware allocation in a way that keeps costs low and results high. Low-power consumption, security, and efficient data transfer from the edge sensors to the edge microcontrollers are critical for ML optimization. This article is a technical perspective and discussion of the current state of artificial intelligence (AI) and machine learning (ML), articulating how Edge AI is penetrating the markets of today and tomorrow, and how it will positively impact our lives in the coming years.