Dens-PU: PU Learning with Density-Based Positive Labeled Augmentation
This addresses PU learning, a semi-supervised classification challenge where only positive and unlabeled data are available, with incremental improvements for image data applications.
The study tackled the PU learning problem by augmenting positive-labeled data using density-based anomaly detection to define a class boundary, achieving state-of-the-art results on benchmark image datasets.
This study proposes a novel approach for solving the PU learning problem based on an anomaly-detection strategy. Latent encodings extracted from positive-labeled data are linearly combined to acquire new samples. These new samples are used as embeddings to increase the density of positive-labeled data and, thus, define a boundary that approximates the positive class. The further a sample is from the boundary the more it is considered as a negative sample. Once a set of negative samples is obtained, the PU learning problem reduces to binary classification. The approach, named Dens-PU due to its reliance on the density of positive-labeled data, was evaluated using benchmark image datasets, and state-of-the-art results were attained.