Contrastive Active Adaptive Partial Label Learning under Class Distribution Mismatch

Abstract

Abstract Partial label learning is an important learning framework where each training sample is associated with a candidate label set and its ground-truth label is included in the candidate label set. Semi-supervised partial label learning is defined as a special case of partial label learning problem where the training sample set consists of partially labeled sample subset and unlabeled sample subset. However, there exists the problem of class distribution mismatch, wherein the unlabeled sample set contains many instances out of the target categories. In this paper, we propose a contrastive active adaptive partial label learning method which combines the active partial learning with the contrastive coding. A novel active sample selection strategy is established to use label propagation ability to measure the optimization ability of unlabeled samples to partially labeled samples. Furthermore, to solve the problem of class distribution mismatch, a jointly query score based on contrastive coding is utilized to reduce the queries of unlabeled samples out of target categories. Finally, the above two indicators are combined adaptively to select the most valuable unlabeled samples in target categories for manual labeling and the selected samples will be added to the training sample set to train the new classifier. The performance and effectiveness of our method is evaluated by performing experiments on the actual data set CIFAR10.