ML Anthology

ARC-LH: A New Adaptive Resampling Algorithm for Improving ANN Classifiers

Friedrich Leisch, Kurt Hornik
NeurIPS 1996 pp. 522-528
/neurips/1996/leisch1996neurips-arclh/

Abstract

We introduce arc-Ih, a new algorithm for improvement of ANN clas(cid:173) sifier performance, which measures the importance of patterns by aggregated network output errors. On several artificial benchmark problems, this algorithm compares favorably with other resample and combine techniques.

PDF NeurIPS Semantic Scholar

Cite

Text

Leisch and Hornik. "ARC-LH: A New Adaptive Resampling Algorithm for Improving ANN Classifiers." Neural Information Processing Systems, 1996.

Markdown

[Leisch and Hornik. "ARC-LH: A New Adaptive Resampling Algorithm for Improving ANN Classifiers." Neural Information Processing Systems, 1996.](https://mlanthology.org/neurips/1996/leisch1996neurips-arclh/)

BibTeX

@inproceedings{leisch1996neurips-arclh,
  title     = {{ARC-LH: A New Adaptive Resampling Algorithm for Improving ANN Classifiers}},
  author    = {Leisch, Friedrich and Hornik, Kurt},
  booktitle = {Neural Information Processing Systems},
  year      = {1996},
  pages     = {522-528},
  url       = {https://mlanthology.org/neurips/1996/leisch1996neurips-arclh/}
}