Model maintenance based on domain adaptation and transfer learning

In this course, you’ll learn how to maintain machine learning models in the face of changing data conditions. You’ll explore different types of dataset and prediction shifts, understand how they affect model performance, and discover practical techniques for detecting and addressing these shifts. Through clear explanations and applied examples, the course introduces key metrics and methods for adapting models using various levels of labelled data. 

What you will learn: 

• How to identify and differentiate causes of model degradation. 
• How to interpret monitoring metrics for dataset and prediction shifts. 
• How to update and maintain models using unlabelled, partially labelled, and fully labelled data. 
• How to apply shift detection and model correction strategies in real-world scenarios. 

Follow the ‘go to course’ and sign up!