Using Unlabeled Data for Supervised Learning

Abstract

Many classification problems have the property that the only costly part of obtaining examples is the class label. This paper suggests a simple method for using distribution information contained in unlabeled examples to augment labeled examples in a supervised training framework. Empirical tests show that the technique de(cid:173) scribed in this paper can significantly improve the accuracy of a supervised learner when the learner is well below its asymptotic accuracy level.

Cite

Text

Towell. "Using Unlabeled Data for Supervised Learning." Neural Information Processing Systems, 1995.

Markdown

[Towell. "Using Unlabeled Data for Supervised Learning." Neural Information Processing Systems, 1995.](https://mlanthology.org/neurips/1995/towell1995neurips-using/)

BibTeX

@inproceedings{towell1995neurips-using,
  title     = {{Using Unlabeled Data for Supervised Learning}},
  author    = {Towell, Geoffrey G.},
  booktitle = {Neural Information Processing Systems},
  year      = {1995},
  pages     = {647-653},
  url       = {https://mlanthology.org/neurips/1995/towell1995neurips-using/}
}