ExactBoost: Directly Boosting the Margin in Combinatorial and Non-Decomposable Metrics

Abstract

Many classification algorithms require the use of surrogate losses when the intended loss function is combinatorial or non-decomposable. This paper introduces a fast and exact stagewise optimization algorithm, dubbed ExactBoost, that boosts stumps to the actual loss function. By developing a novel extension of margin theory to the non-decomposable setting, it is possible to provably bound the generalization error of ExactBoost for many important metrics with different levels of non-decomposability. Through extensive examples, it is shown that such theoretical guarantees translate to competitive empirical performance. In particular, when used as an ensembler, ExactBoost is able to significantly outperform other surrogate-based and exact algorithms available.

Cite

Text

Csillag et al. " ExactBoost: Directly Boosting the Margin in Combinatorial and Non-Decomposable Metrics ." Artificial Intelligence and Statistics, 2022.

Markdown

[Csillag et al. " ExactBoost: Directly Boosting the Margin in Combinatorial and Non-Decomposable Metrics ." Artificial Intelligence and Statistics, 2022.](https://mlanthology.org/aistats/2022/csillag2022aistats-exactboost/)

BibTeX

@inproceedings{csillag2022aistats-exactboost,
  title     = {{ ExactBoost: Directly Boosting the Margin in Combinatorial and Non-Decomposable Metrics }},
  author    = {Csillag, Daniel and Piazza, Carolina and Ramos, Thiago and Vitor Romano, João and Oliveira, Roberto I. and Orenstein, Paulo},
  booktitle = {Artificial Intelligence and Statistics},
  year      = {2022},
  pages     = {9017-9049},
  volume    = {151},
  url       = {https://mlanthology.org/aistats/2022/csillag2022aistats-exactboost/}
}