Balanced Self-Paced Learning for AUC Maximization

Authors

  • Bin Gu Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), Abu Dhabi, UAE Nanjing University of Information Science & Technology, Nanjing, China
  • Chenkang Zhang Nanjing University of Information Science & Technology, Nanjing, China
  • Huan Xiong Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), Abu Dhabi, UAE Institute for Advanced Study in Mathematics, Harbin Institute of Technology, China
  • Heng Huang University of Pittsburgh, Pittsburgh, USA

DOI:

https://doi.org/10.1609/aaai.v36i6.20632

Keywords:

Machine Learning (ML)

Abstract

Learning to improve AUC performance is an important topic in machine learning. However, AUC maximization algorithms may decrease generalization performance due to the noisy data. Self-paced learning is an effective method for handling noisy data. However, existing self-paced learning methods are limited to pointwise learning, while AUC maximization is a pairwise learning problem. To solve this challenging problem, we innovatively propose a balanced self-paced AUC maximization algorithm (BSPAUC). Specifically, we first provide a statistical objective for self-paced AUC. Based on this, we propose our self-paced AUC maximization formulation, where a novel balanced self-paced regularization term is embedded to ensure that the selected positive and negative samples have proper proportions. Specially, the sub-problem with respect to all weight variables may be non-convex in our formulation, while the one is normally convex in existing self-paced problems. To address this, we propose a doubly cyclic block coordinate descent method. More importantly, we prove that the sub-problem with respect to all weight variables converges to a stationary point on the basis of closed-form solutions, and our BSPAUC converges to a stationary point of our fixed optimization objective under a mild assumption. Considering both the deep learning and kernel-based implementations, experimental results on several large-scale datasets demonstrate that our BSPAUC has a better generalization performance than existing state-of-the-art AUC maximization methods.

Downloads

Published

2022-06-28

How to Cite

Gu, B., Zhang, C., Xiong, H., & Huang, H. (2022). Balanced Self-Paced Learning for AUC Maximization. Proceedings of the AAAI Conference on Artificial Intelligence, 36(6), 6765-6773. https://doi.org/10.1609/aaai.v36i6.20632

Issue

Section

AAAI Technical Track on Machine Learning I