@article{Ni2021, 
author = {Peng Ni and Su-Yun Zhao and Zhi-Gang Dai and Hong Chen and Cui-Ping Li},
title = {Partial Label Learning via Conditional-Label-Aware Disambiguation},
year = {2021},
journal = {Journal of Computer Science and Technology},
volume = {36},
number = {3},
pages = {590-605},
keywords = {disambiguation, partial label learning, similarity and dissimilarity, weak supervision},
url = {https://www.sciopen.com/article/10.1007/s11390-021-0992-x},
doi = {10.1007/s11390-021-0992-x},
abstract = {Partial label learning is a weakly supervised learning framework in which each instance is associated with multiple candidate labels, among which only one is the ground-truth label. This paper proposes a unified formulation that employs proper label constraints for training models while simultaneously performing pseudo-labeling. Unlike existing partial label learning approaches that only leverage similarities in the feature space without utilizing label constraints, our pseudo-labeling process leverages similarities and differences in the feature space using the same candidate label constraints and then disambiguates noise labels. Extensive experiments on artificial and real-world partial label datasets show that our approach significantly outperforms state-of-the-art counterparts on classification prediction.}
}