ML Anthology

No Free Lunch for Early Stopping

Zehra Cataltepe, Yaser S. Abu-Mostafa, Malik Magdon-Ismail
NeCo 1999 pp. 995-1009
doi:10.1162/089976699300016557 /neco/1999/cataltepe1999neco-free/

Abstract

We show that with a uniform prior on models having the same training error, early stopping at some fixed training error above the training error minimum results in an increase in the expected generalization error.

PDF NeCo Semantic Scholar

Cite

Text

Cataltepe et al. "No Free Lunch for Early Stopping." Neural Computation, 1999. doi:10.1162/089976699300016557

Markdown

[Cataltepe et al. "No Free Lunch for Early Stopping." Neural Computation, 1999.](https://mlanthology.org/neco/1999/cataltepe1999neco-free/) doi:10.1162/089976699300016557

BibTeX

@article{cataltepe1999neco-free,
  title     = {{No Free Lunch for Early Stopping}},
  author    = {Cataltepe, Zehra and Abu-Mostafa, Yaser S. and Magdon-Ismail, Malik},
  journal   = {Neural Computation},
  year      = {1999},
  pages     = {995-1009},
  doi       = {10.1162/089976699300016557},
  volume    = {11},
  url       = {https://mlanthology.org/neco/1999/cataltepe1999neco-free/}
}