Watch a 59-minute seminar from the Federated Learning One World series featuring MIT's Song Han as he explores the challenges and solutions for on-device training with extremely limited memory constraints of 256KB. Learn about innovative approaches to machine learning deployment on resource-constrained devices, drawing from Han's expertise in efficient AI systems and tiny machine learning. Discover practical techniques for implementing neural network training in environments where memory is at a premium, making AI accessible on edge devices and microcontrollers.
Overview
Syllabus
FLOW Seminar #108: Song Han (MIT) On-Device Training under 256KB of Memory
Taught by
Federated Learning One World Seminar
Related Courses
-
Getting Started with Machine Learning at the Edge on Arm
-
On-Device Training and Transfer Learning - Lecture 15
-
On-Device Training and Transfer Learning - Lecture 15
-
On-Device Learning Under 256KB Memory - Challenges and Solutions for IoT Devices
-
Tiny Machine Learning (TinyML)
-
MCUNet and TinyML - Lecture 10
