ML Anthology

Semi-Supervised Learning of Class Balance Under Class-Prior Change by Distribution Matching

Marthinus Christoffel du Plessis, Masashi Sugiyama
ICML 2012
doi:10.1016/j.neunet.2013.11.010 /icml/2012/duplessis2012icml-semi/

Abstract

In real-world classification problems, the class balance in the training dataset does not necessarily reflect that of the test dataset, which can cause significant estimation bias. If the class ratio of the test dataset is known, instance re-weighting or resampling allows systematical bias correction. However, learning the class ratio of the test dataset is challenging when no labeled data is available from the test domain. In this paper, we propose to estimate the class ratio in the test dataset by matching probability distributions of training and test input data. We demonstrate the utility of the proposed approach through experiments.

PDF Semantic Scholar

Cite

Text

du Plessis and Sugiyama. "Semi-Supervised Learning of Class Balance Under Class-Prior Change by Distribution Matching." International Conference on Machine Learning, 2012. doi:10.1016/j.neunet.2013.11.010

Markdown

[du Plessis and Sugiyama. "Semi-Supervised Learning of Class Balance Under Class-Prior Change by Distribution Matching." International Conference on Machine Learning, 2012.](https://mlanthology.org/icml/2012/duplessis2012icml-semi/) doi:10.1016/j.neunet.2013.11.010

BibTeX

@inproceedings{duplessis2012icml-semi,
  title     = {{Semi-Supervised Learning of Class Balance Under Class-Prior Change by Distribution Matching}},
  author    = {du Plessis, Marthinus Christoffel and Sugiyama, Masashi},
  booktitle = {International Conference on Machine Learning},
  year      = {2012},
  doi       = {10.1016/j.neunet.2013.11.010},
  url       = {https://mlanthology.org/icml/2012/duplessis2012icml-semi/}
}