Guiding CNNs towards Relevant Concepts by Multi-task and Adversarial Learning

by   Mara Graziani, et al.

The opaqueness of deep learning limits its deployment in critical application scenarios such as cancer grading in medical images. In this paper, a framework for guiding CNN training is built on top of successful existing techniques of hard parameter sharing, with the main goal of explicitly introducing expert knowledge in the training objectives. The learning process is guided by identifying concepts that are relevant or misleading for the task. Relevant concepts are encouraged to appear in the representation through multi-task learning. Undesired and misleading concepts are discouraged by a gradient reversal operation. In this way, a shift in the deep representations can be corrected to match the clinicians' assumptions. The application on breast lymph nodes histopathology data from the Camelyon challenge shows a significant increase in the generalization performance on unseen patients (from 0.839 to 0.864 average AUC, p-value = 0,0002) when the internal representations are controlled by prior knowledge regarding the acquisition center and visual features of the tissue. The code will be shared for reproducibility on our GitHub repository.


page 1

page 2

page 3

page 4


Cross-connected Networks for Multi-task Learning of Detection and Segmentation

Multi-task learning improves generalization performance by sharing knowl...

On the relationship between disentanglement and multi-task learning

One of the main arguments behind studying disentangled representations i...

Representation Disentanglement for Multi-task Learning with application to Fetal Ultrasound

One of the biggest challenges for deep learning algorithms in medical im...

Multi-Task Learning by Deep Collaboration and Application in Facial Landmark Detection

Convolutional neural networks (CNN) have become the most successful and ...

To Reverse the Gradient or Not: An Empirical Comparison of Adversarial and Multi-task Learning in Speech Recognition

Transcribed datasets typically contain speaker identity for each instanc...

Internal-transfer Weighting of Multi-task Learning for Lung Cancer Detection

Recently, multi-task networks have shown to both offer additional estima...

Please sign up or login with your details

Forgot password? Click here to reset