Many innovations in today’s edge technology began with consumer electronics and mobile devices. The smartphones, tablets, cameras, smart speakers, drones, and other devices we use in our daily lives are actually various forms of AI edge devices.
These devices are embedded with many different operating systems and optimizations for specific chips to enable them to run specific use cases. For example, drones typically come with built-in cameras, and sometimes they even include energy-efficient processors to extend their runtime and battery life.
Recently, Adam Gibson, the Chief Technology Officer of Skymind and the CEO and co-founder of Konduit (a subsidiary of Skymind), appeared at the World Artificial Intelligence Conference Developer Day’s enterprise AI architecture session, delivering a keynote speech titled ‘TinyML and AI on Edge Devices: A New Paradigm for Ubiquitous AI’ to online audiences.
It is recommended to watch in a WIFI environment
According to a survey by Eclipse IoT, most people still do not consider machine learning; they are more focused on the types of sensors they can implement and the types of data they can capture. Gibson stated, ‘While most people or companies have ideas about machine learning, understanding the currently available machine learning cases has not been prioritized. I believe we will see this trend increase in the next 3 to 5 years.’
TinyML typically refers to deep learning frameworks specifically designed to run on low-power devices, with many innovations aimed at how deep learning can operate on smaller devices. Just being able to run analytics does not mean it can be efficient. This is why we emphasize low-power microcontrollers.
To learn more about the technology and applications of TinyML and AI on edge devices, please click above to watch the speech transcript.
Long press to scan the QR code to follow us
Open the door to the world of artificial intelligence
