TinyML and efficient deep learning
Have you found it difficult to deploy neural networks on mobile devices and IoT devices? Have you ever found it too slow to train neural networks? Presented by Prof. Song Han, MIT EECS, this course is a deep dive into efficient machine learning techniques that enable powerful deep learning applications on resource-constrained devices.
Topics covered
- Efficient inference techniques, including model compression, pruning, quantization, neural architecture search, and distillation.
- Efficient training techniques, including gradient compression and on-device transfer learning.
- Application-specific model optimization techniques for videos, point cloud, and NLP; and efficient quantum machine learning.
The 32F746GDISCOVERY Discovery kit is used in the on-training device of the course.
Ready to get started?
The entire course is available on YouTube at the URL: https://www.youtube.com/playlist?list=PL80kAHvQbh-pT4lCkDT53zT8DKmhE0idB
Join the discussion
Our MIT representative (username) is ready to answer your questions regarding this curriculum. Feel free to ask and share your thoughts.