Wasserstein Adversarial Regularization (WAR) on label noise

Bharath Damodaran, Kilian Fatras, Sylvain Lobry, Rémi Flamary, Devis Tuia, Nicolas Courty

Keywords: Label Noise, Adversarial regularization, Wasserstein

TL;DR: We present a novel method for handling label noise through an adversarial regularization incorporating a Wasserstein distance

Abstract:

Noisy labels often occur in vision datasets, especially when they are obtained from crowdsourcing or Web scraping. We propose a new regularization method, which enables learning robust classifiers in presence of noisy data. To achieve this goal, we propose a new adversarial regularization scheme based on the Wasserstein distance. Using this distance allows taking into account specific relations between classes by leveraging the geometric properties of the labels space. Our Wasserstein Adversarial Regularization (WAR) encodes a selective regularization, which promotes smoothness of the classifier between some classes, while preserving sufficient complexity of the decision boundary between others. We first discuss how and why adversarial regularization can be used in the context of label noise and then show the effectiveness of our method on five datasets corrupted with noisy labels: in both benchmarks and real datasets, WAR outperforms the state-of-the-art
competitors.