A Comparative Study of Deep Learning Loss Functions for Multi-Label Remote Sensing Image Classification

TL;DR

This study compares seven deep learning loss functions for multi-label remote sensing image classification, providing theoretical and experimental insights to guide loss function selection based on accuracy, class imbalance, and efficiency.

Contribution

It is the first comprehensive analysis of various loss functions in remote sensing, offering guidelines for their effective application in multi-label classification tasks.

Findings

Focal loss improves class imbalance handling.

Weighted cross-entropy enhances accuracy in imbalanced datasets.

Ranking loss shows faster convergence.

Abstract

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.