Learning Safe Prediction for Semi-Supervised Regression

Authors

  • Yu-Feng Li Nanjing University
  • Han-Wen Zha University of California, Santa Barbara
  • Zhi-Hua Zhou Nanjing University

DOI:

https://doi.org/10.1609/aaai.v31i1.10856

Keywords:

Semi-supervised regression, Safe

Abstract

Semi-supervised learning (SSL) concerns how to improve performance via the usage of unlabeled data. Recent studies indicate that the usage of unlabeled data might even deteriorate performance. Although some proposals have been developed to alleviate such a fundamental challenge for semi-supervised classification, the efforts on semi-supervised regression (SSR) remain to be limited. In this work we consider the learning of a safe prediction from multiple semi-supervised regressors, which is not worse than a direct supervised learner with only labeled data. We cast it as a geometric projection issue with an efficient algorithm. Furthermore, we show that the proposal is provably safe and has already achieved the maximal performance gain, if the ground-truth label assignment is realized by a convex linear combination of base regressors. This provides insight to help understand safe SSR. Experimental results on a broad range of datasets validate the effectiveness of our proposal.

Downloads

Published

2017-02-13

How to Cite

Li, Y.-F., Zha, H.-W., & Zhou, Z.-H. (2017). Learning Safe Prediction for Semi-Supervised Regression. Proceedings of the AAAI Conference on Artificial Intelligence, 31(1). https://doi.org/10.1609/aaai.v31i1.10856