On-Device Training of Machine Learning Models with MATLAB and Simulink

Learn how to use MATLAB and Simulink to create machine learning models that can be trained on an embedded device to adapt to new data.

We'll cover:

The basics of on-device learning techniques and typical applications and devices

Motivation for training machine learning models on embedded devices

Challenges involved in on-device learning

Two main approaches to on-device learning: using passive or active model updates

The steps in a typical workflow using an audio classification example

About the Presenters:

Jack Ferrari is a product manager at MathWorks, focused on code generation for deep learning models. He is also the product manager for Deep Learning Toolbox Model Quantization Library, enabling MATLAB users to compress and deploy AI models to edge devices and embedded systems. Jack holds a B.S. in Mechanical Engineering from Boston University.

Brad Duncan is the product manager for machine learning capabilities in the Statistics and Machine Learning Toolbox at MathWorks. He works with customers to apply AI in new areas of engineering such as incorporating virtual sensors in engineered systems, building explainable machine learning models, and standardizing AI workflows using MATLAB and Simulink.