Going Beyond One-Hot Encoding in Classification: Can Human Uncertainty Improve Model Performance in Earth Observation?

• Published in: IEEE transactions on geoscience and remote sensing Vol. 62; p. 1
• Main Authors: Koller, Christoph, Kauermann, Goran, Zhu, Xiao Xiang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Calibration; Classification; Deep learning; Human Uncertainty; Labels; Local Climate Zones (LCZ); Low density lipoprotein; Machine learning; Meteorology; Predictive models; Remote sensing; Subject specialists; Task analysis; Training; Uncertainty; Uncertainty Quantification; Urban areas
• ISSN: 0196-2892; 1558-0644
• DOI: 10.1109/TGRS.2023.3336357

Technological and computational advances continuously drive forward the field of deep learning in remote sensing. In recent years, the derivation of quantities describing the uncertainty in the prediction - which naturally accompanies the modeling process - has sparked interest in the remote sensing community. Often neglected in the machine learning setting is the human uncertainty that influences numerous labeling processes. As the core of this work, the task of Local Climate Zone (LCZ) classification is studied by means of a data set that contains multiple label votes by domain experts for each image. The inherent label uncertainty describes the ambiguity among the domain experts and is explicitly embedded into the training process via distributional labels. We show that incorporating the label uncertainty helps the model to generalize better to the test data and increases model performance. Similar to existing calibration methods, the distributional labels lead to better-calibrated probabilities, which in turn yield more certain and trustworthy predictions. For reproducibility, we provide our code here https://github.com/ChrisKo94/LCZ_LDL and here https://gitlab.lrz.de/ai4eo/WG_Uncertainty/lcz_ldl.