ML Anthology

ABC-Boost: Adaptive Base Class Boost for Multi-Class Classification

Ping Li
ICML 2009 pp. 625-632
doi:10.1145/1553374.1553455 /icml/2009/li2009icml-abc/

Abstract

We propose abc-boost (adaptive base class boost) for multi-class classification and present abc-mart, an implementation of abc-boost, based on the multinomial logit model. The key idea is that, at each boosting iteration, we adaptively and greedily choose a base class. Our experiments on public datasets demonstrate the improvement of abc-mart over the original mart algorithm.

ICML Semantic Scholar

Cite

Text

Li. "ABC-Boost: Adaptive Base Class Boost for Multi-Class Classification." International Conference on Machine Learning, 2009. doi:10.1145/1553374.1553455

Markdown

[Li. "ABC-Boost: Adaptive Base Class Boost for Multi-Class Classification." International Conference on Machine Learning, 2009.](https://mlanthology.org/icml/2009/li2009icml-abc/) doi:10.1145/1553374.1553455

BibTeX

@inproceedings{li2009icml-abc,
  title     = {{ABC-Boost: Adaptive Base Class Boost for Multi-Class Classification}},
  author    = {Li, Ping},
  booktitle = {International Conference on Machine Learning},
  year      = {2009},
  pages     = {625-632},
  doi       = {10.1145/1553374.1553455},
  url       = {https://mlanthology.org/icml/2009/li2009icml-abc/}
}