A Generalized Unbiased Risk Estimator for Learning with Augmented Classes


  • Senlin Shu Chongqing University
  • Shuo He University of Electronic Science and Technology of China
  • Haobo Wang Zhejiang University
  • Hongxin Wei Nanyang Technological University
  • Tao Xiang Chongqing University
  • Lei Feng Nanyang Technological University




ML: Multi-Class/Multi-Label Learning & Extreme Classification, ML: Classification and Regression, ML: Semi-Supervised Learning


In contrast to the standard learning paradigm where all classes can be observed in training data, learning with augmented classes (LAC) tackles the problem where augmented classes unobserved in the training data may emerge in the test phase. Previous research showed that given unlabeled data, an unbiased risk estimator (URE) can be derived, which can be minimized for LAC with theoretical guarantees. However, this URE is only restricted to the specific type of one-versus-rest loss functions for multi-class classification, making it not flexible enough when the loss needs to be changed with the dataset in practice. In this paper, we propose a generalized URE that can be equipped with arbitrary loss functions while maintaining the theoretical guarantees, given unlabeled data for LAC. To alleviate the issue of negative empirical risk commonly encountered by previous studies, we further propose a novel risk-penalty regularization term. Experiments demonstrate the effectiveness of our proposed method.




How to Cite

Shu, S., He, S., Wang, H., Wei, H., Xiang, T., & Feng, L. (2023). A Generalized Unbiased Risk Estimator for Learning with Augmented Classes. Proceedings of the AAAI Conference on Artificial Intelligence, 37(8), 9829-9836. https://doi.org/10.1609/aaai.v37i8.26173



AAAI Technical Track on Machine Learning III