Positive unlabeled learning with tensor networks
This addresses a common issue in domains like medicine and advertising where negative labels are hard to obtain, offering a non-domain-specific solution with generative capabilities.
The paper tackles the positive unlabeled learning problem, where only positive and unlabeled data are available, by introducing a tensor network approach that improves state-of-the-art results on MNIST and 15 categorical/mixed datasets, and enables generation of new positive and negative instances.
Positive unlabeled learning is a binary classification problem with positive and unlabeled data. It is common in domains where negative labels are costly or impossible to obtain, e.g., medicine and personalized advertising. Most approaches to positive unlabeled learning apply to specific data types (e.g., images, categorical data) and can not generate new positive and negative samples. This work introduces a feature-space distance-based tensor network approach to the positive unlabeled learning problem. The presented method is not domain specific and significantly improves the state-of-the-art results on the MNIST image and 15 categorical/mixed datasets. The trained tensor network model is also a generative model and enables the generation of new positive and negative instances.