On-Device Training and Transfer Learning - Lecture 16

Explore on-device training and transfer learning techniques in this lecture from MIT's course on TinyML and Efficient Deep Learning Computing. Dive into system support for efficient on-device training, focusing on the Tiny Training Engine. Gain insights into deploying neural networks on mobile and IoT devices, and learn strategies for overcoming challenges in training speed. Examine topics such as model compression, pruning, quantization, neural architecture search, and distillation for efficient inference. Discover efficient training techniques including gradient compression and on-device transfer learning. Investigate application-specific model optimization for videos, point cloud, and NLP, as well as efficient quantum machine learning. Access accompanying slides and course materials to enhance your understanding of these cutting-edge concepts in machine learning efficiency.

Lecture 16 - On-Device Training and Transfer Learning (Part II) | MIT 6.S965