Predictive Adversarial Learning from Positive and Unlabeled Data

Authors

  • Wenpeng Hu Peking University
  • Ran Le Peking University
  • Bing Liu UIC
  • Feng Ji Alibaba Inc.
  • Jinwen Ma Peking University
  • Dongyan Zhao Peking University
  • Rui Yan Peking University

Keywords:

Classification and Regression

Abstract

This paper studies learning from positive and unlabeled examples, known as PU learning. It proposes a novel PU learning method called Predictive Adversarial Networks (PAN) based on GAN (Generative Adversarial Networks). GAN learns a generator to generate data (e.g., images) to fool a discriminator which tries to determine whether the generated data belong to a (positive) training class. PU learning can be casted as trying to identify (not generate) likely positive instances from the unlabeled set to fool a discriminator that determines whether the identified likely positive instances from the unlabeled set are indeed positive. However, directly applying GAN is problematic because GAN focuses on only the positive data. The resulting PU learning method will have high precision but low recall. We propose a new objective function based on KL-divergence. Evaluation using both image and text data shows that PAN outperforms state-of-the-art PU learning methods and also a direct adaptation of GAN for PU learning.

Downloads

Published

2021-05-18

How to Cite

Hu, W., Le, R., Liu, B., Ji, F., Ma, J., Zhao, D., & Yan, R. (2021). Predictive Adversarial Learning from Positive and Unlabeled Data. Proceedings of the AAAI Conference on Artificial Intelligence, 35(9), 7806-7814. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/16953

Issue

Section

AAAI Technical Track on Machine Learning II