Explore the world of ultra-low power machine learning systems through this 24-minute conference talk from the tinyML Summit 2020. Delve into the concept of tinyMLPerf, a benchmarking tool for tiny machine learning systems, as presented by Vijay Janapa Reddi, MLPerf Inference Chair and Associate Professor at Harvard University. Gain insights into the methodology, ecosystem impact, and unique aspects of tinyML, including quantization and pruning techniques. Understand the community-driven effort behind tinyMLPerf, its set of tasks, and how it enables performance evaluation in the rapidly evolving field of ultra-low power machine learning.
tinyMLPerf: Benchmarking Ultra-low Power Machine Learning Systems
Overview
Syllabus
Introduction
About Vijay
What is tinyML
What is tinyMLPerf
How does tinyMLPerf benchmark
What does tinyMLPerf enable
tinyMLPerf benchmark
how does this translate to the ecosystem
whats interesting about this space
quantization
community driven effort
set of tasks
prune
Methodology
Conclusion
Taught by
tinyML
Related Courses
-
Tiny Machine Learning (TinyML)
-
Applied Tiny Machine Learning (TinyML) for Scale
-
Tiny but Powerful: Hardware for High Performance, Low Power Machine Learning
-
Machine Learning Sensor Certifications: Standards, Ethics, and Implementation Framework
-
Application of Ultra-Low-Power and Resource-Constrained Devices to Improve Health Care Delivery in Resource-Constrained Communities
-
MCUNet and TinyML - Lecture 10
