Partial Label Learning by Entropy Minimization

Xuejun Han (Carleton University).

Partial label learning deals with the problem where each training example is associated with a set of candidate labels, only one of which is assumed to be valid. To learn from such ambiguous labeling information, the critical point is to disambiguate the set of candidate labels, thereby targeting the ground-truth label. By utilizing the nature that only one of the candidate labels is correct, we employ the entropy minimization strategy to push the model making confident predictions of the training data. By doing this, the ground-truth labels are likely to make more contributions to the model training. Finally, comparative experiments on a number of real-world datasets are conducted, clearly demonstrateing the effectiveness of the proposed approach.