MIT researchers have developed a new cutting edge system, dubbed MCUNet, which brings machine learning to microcontrollers. This new approach should significantly enhance the function and security of all types of IoT devices. Normally, pattern-recognition tasks such as deep learning are difficult to run locally on IoT devices. However now deploying neural networks directly on these tiny devices has become a hot new research area. The following article describes the thinking behind how the MIT research team plans to make MCUNet a reality.
Deep learning is everywhere. This branch of artificial intelligence curates your social media and serves your Google search results. Soon, deep learning could also check your vitals or set your thermostat. MIT researchers have developed a system that could bring deep learning neural networks to new — and much smaller — places, like the tiny computer chips in wearable medical devices, household appliances, and the 250 billion other objects that constitute the “internet of things” (IoT).
The system, called MCUNet, designs compact neural networks that deliver unprecedented speed and accuracy for deep learning on IoT devices, despite limited memory and processing power. The technology could facilitate the expansion of the IoT universe while saving energy and improving data security.
The research will be presented at next month’s Conference on Neural Information Processing Systems.
Read the full story on MIT News
