An Entropic Optimal Transport Loss for Learning Deep Neural Networks under Label Noise in Remote Sensing Images

Abstract

Deep neural networks have established as a powerful tool for large scale supervised classification tasks. The state-of-the-art performances of deep neural networks are conditioned to the availability of large number of accurately labeled samples. In practice, collecting large scale accurately labeled datasets is a challenging and tedious task in most scenarios of remote sensing image analysis, thus cheap surrogate procedures are employed to label the dataset. Training deep neural networks on such datasets with inaccurate labels easily overfits to the noisy training labels and degrades the performance of the classification tasks drastically. To mitigate this effect, we propose an original solution with entropic optimal transportation. It allows to learn in an end-to-end fashion deep neural networks that are, to some extent, robust to inaccurately labeled samples. We empirically demonstrate on several remote sensing datasets, where both scene and pixel-based hyperspectral images are considered for classification. Our method proves to be highly tolerant to significant amounts of label noise and achieves favorable results against state-of-the-art methods.

Under Consideration at Computer Vision and Image Understanding

References in corpus (10)

• Understanding deep learning requires rethinking generalization
• Training Deep Neural Networks on Noisy Labels with Bootstrapping
• Training Convolutional Networks with Noisy Labels
• Detecting Mammals in UAV Images: Best Practices to address a substantially Imbalanced Dataset with Deep Learning
• Learning Aerial Image Segmentation from Online Maps
• Robust Loss Functions under Label Noise for Deep Neural Networks
• Toward Robustness against Label Noise in Training Deep Discriminative Neural Networks
• Differential Properties of Sinkhorn Approximation for Learning with Wasserstein Distance
• High-Resolution Multispectral Dataset for Semantic Segmentation
• TorontoCity: Seeing the World with a Million Eyes

Cited by in corpus (4)

• On the Effects of Different Types of Label Noise in Multi-Label Remote Sensing Image Classification
• Sinkhorn Divergences for Unbalanced Optimal Transport
• Unbalanced minibatch Optimal Transport; applications to Domain Adaptation
• A Novel Technique for Robust Training of Deep Networks With Multisource Weak Labeled Remote Sensing Data