Back to All Events
Presenter: Weiguang Guan, SHARCNET
Contrastive learning is a machine learning technique used to learn a representation of the input data that maximizes the difference between samples of different classes and minimizes the difference between samples of the same class. The learned representation (or features) will then be used to solve a classification problem. In this tutorial, we show this effective learning technique from head to toe through an image classification example. As you can see, contrastive learning plays a role of feature extractor which helps subsequent classification training to achieve higher accuracy.