A Comparative Study of Deep Learning Loss Functions for Multi-Label Remote Sensing Image Classification
It addresses the problem of optimizing loss function choice for researchers and practitioners in remote sensing, but it is incremental as it applies existing methods to a new domain.
This paper compared seven deep learning loss functions for multi-label remote sensing image classification, analyzing their performance in accuracy, class imbalance handling, and efficiency, and provided guidelines for selecting appropriate loss functions in this domain.
This paper analyzes and compares different deep learning loss functions in the framework of multi-label remote sensing (RS) image scene classification problems. We consider seven loss functions: 1) cross-entropy loss; 2) focal loss; 3) weighted cross-entropy loss; 4) Hamming loss; 5) Huber loss; 6) ranking loss; and 7) sparseMax loss. All the considered loss functions are analyzed for the first time in RS. After a theoretical analysis, an experimental analysis is carried out to compare the considered loss functions in terms of their: 1) overall accuracy; 2) class imbalance awareness (for which the number of samples associated to each class significantly varies); 3) convexibility and differentiability; and 4) learning efficiency (i.e., convergence speed). On the basis of our analysis, some guidelines are derived for a proper selection of a loss function in multi-label RS scene classification problems.