Can Class-Priors Help Single-Positive Multi-Label Learning?
This work addresses a weakly supervised learning problem for multi-label classification, offering an incremental improvement by integrating class-priors into the SPMLL framework.
The paper tackles the problem of single-positive multi-label learning (SPMLL), where training examples have only one annotated positive label, by proposing a framework that incorporates class-priors to address the unrealistic assumption of identical prior probabilities, resulting in improved performance over existing methods on ten benchmark datasets.
Single-positive multi-label learning (SPMLL) is a typical weakly supervised multi-label learning problem, where each training example is annotated with only one positive label. Existing SPMLL methods typically assign pseudo-labels to unannotated labels with the assumption that prior probabilities of all classes are identical. However, the class-prior of each category may differ significantly in real-world scenarios, which makes the predictive model not perform as well as expected due to the unrealistic assumption on real-world application. To alleviate this issue, a novel framework named {\proposed}, i.e., Class-pRiors Induced Single-Positive multi-label learning, is proposed. Specifically, a class-priors estimator is introduced, which could estimate the class-priors that are theoretically guaranteed to converge to the ground-truth class-priors. In addition, based on the estimated class-priors, an unbiased risk estimator for classification is derived, and the corresponding risk minimizer could be guaranteed to approximately converge to the optimal risk minimizer on fully supervised data. Experimental results on ten MLL benchmark datasets demonstrate the effectiveness and superiority of our method over existing SPMLL approaches.