Abstract:
Edge Machine Learning (EdgeML) is known as a rapidly expanding field of machine learning technologies and applications that encompasses hardware, software, and algorithms. But are we, as businesses and academia, ready for this widespread emerging technology? Machine learning has been around for a long time and has a wide range of applications. It has, however, been less commonly linked to hardware. Machine learning and hardware used to be linked to the cloud, which typically is associated with latency, power consumption, and putting machines at the mercy of connection speeds and the risk of security. Applying ML to embedded devices is not new, and most of us already have some form of neural network on our phones. However, there are many challenges and concerns when it comes to the embedded ML, some of which are power and space. That’s where EdgeML comes in and brings machine learning to the scene and makes it capable of performing on-device sensor data analytics at extremely low power, typically in the mW range and below, and hence enabling a variety of always-on use-cases and targeting battery-operated devices. According to chip giant Intel, adopting edge technologies will be key to businesses’ success. However, moving ML to the edge has critical requirements and challenges from different points of view. In this talk, we are going to dive into different aspects of EdgeML and discuss its existing HW, frameworks, platforms, and market opportunities. This will help us to see the current status as well as the future of EdgeML and indicate our readiness for it.
Online Recording: https://youtu.be/AWBtGcYHYaM
