The Effect of the Loss on Generalization: Empirical Study on Synthetic Lung Nodule Data
This provides insights for designing deep learning solutions in medical imaging, but it is incremental as it builds on existing contrastive learning and generalization research.
The study investigated how different loss functions affect the features learned by CNNs and their generalization on out-of-distribution data, using a synthetic lung nodule dataset, finding that loss choice influences feature learning and generalization ability.
Convolutional Neural Networks (CNNs) are widely used for image classification in a variety of fields, including medical imaging. While most studies deploy cross-entropy as the loss function in such tasks, a growing number of approaches have turned to a family of contrastive learning-based losses. Even though performance metrics such as accuracy, sensitivity and specificity are regularly used for the evaluation of CNN classifiers, the features that these classifiers actually learn are rarely identified and their effect on the classification performance on out-of-distribution test samples is insufficiently explored. In this paper, motivated by the real-world task of lung nodule classification, we investigate the features that a CNN learns when trained and tested on different distributions of a synthetic dataset with controlled modes of variation. We show that different loss functions lead to different features being learned and consequently affect the generalization ability of the classifier on unseen data. This study provides some important insights into the design of deep learning solutions for medical imaging tasks.