ML Anthology

Randomizing Outputs to Increase Prediction Accuracy

Leo Breiman
MLJ 2000 pp. 229-242
doi:10.1023/A:1007682208299 /mlj/2000/breiman2000mlj-randomizing/

Abstract

Bagging and boosting reduce error by changing both the inputs and outputs to form perturbed training sets, growing predictors on these perturbed training sets and combining them. An interesting question is whether it is possible to get comparable performance by perturbing the outputs alone. Two methods of randomizing outputs are experimented with. One is called output smearing and the other output flipping. Both are shown to consistently do better than bagging.

PDF MLJ Semantic Scholar

Cite

Text

Breiman. "Randomizing Outputs to Increase Prediction Accuracy." Machine Learning, 2000. doi:10.1023/A:1007682208299

Markdown

[Breiman. "Randomizing Outputs to Increase Prediction Accuracy." Machine Learning, 2000.](https://mlanthology.org/mlj/2000/breiman2000mlj-randomizing/) doi:10.1023/A:1007682208299

BibTeX

@article{breiman2000mlj-randomizing,
  title     = {{Randomizing Outputs to Increase Prediction Accuracy}},
  author    = {Breiman, Leo},
  journal   = {Machine Learning},
  year      = {2000},
  pages     = {229-242},
  doi       = {10.1023/A:1007682208299},
  volume    = {40},
  url       = {https://mlanthology.org/mlj/2000/breiman2000mlj-randomizing/}
}