Attention De-sparsification Matters: Inducing diversity in digital pathology representation learning.

Abstract excerpt

We propose DiRL, a Diversity-inducing Representation Learning technique for histopathology imaging. Self-supervised learning (SSL) techniques, such as contrastive and non-contrastive approaches, have been shown to learn rich and effective representations of digitized tissue samples with limited pathologist supervision. Our analysis of vanilla SSL-pretrained models' attention distribution reveals an insightful...

Topics

Share this publication in a Topic to start or enrich a Post.

• Share to Image Processing, Computer-Assisted
• Share to Machine Learning