ICML 2021spotlight10 citations
ChaCha for Online AutoML
Qingyun Wu, Chi Wang, John Langford, Paul Mineiro, Marco Rossi
Abstract
We propose the ChaCha (Champion-Challengers) algorithm for making an online choice of hyperparameters in online learning settings. ChaCha handles the process of determining a champion and scheduling a set of ‘live’ challengers over time based on sample complexity bounds. It is guaranteed to have sublinear regret after the optimal configuration is added into consideration by an application-dependent oracle based on the champions. Empirically, we show that ChaCha provides good performance across a wide array of datasets when optimizing over featurization and hyperparameter decisions.
BibTeX
@InProceedings{pmlr-v139-wu21d,
title = {ChaCha for Online AutoML},
author = {Wu, Qingyun and Wang, Chi and Langford, John and Mineiro, Paul and Rossi, Marco},
booktitle = {Proceedings of the 38th International Conference on Machine Learning},
pages = {11263--11273},
year = {2021},
editor = {Meila, Marina and Zhang, Tong},
volume = {139},
series = {Proceedings of Machine Learning Research},
month = {18--24 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v139/wu21d/wu21d.pdf},
url = {https://proceedings.mlr.press/v139/wu21d.html},
abstract = {We propose the ChaCha (Champion-Challengers) algorithm for making an online choice of hyperparameters in online learning settings. ChaCha handles the process of determining a champion and scheduling a set of ‘live’ challengers over time based on sample complexity bounds. It is guaranteed to have sublinear regret after the optimal configuration is added into consideration by an application-dependent oracle based on the champions. Empirically, we show that ChaCha provides good performance across a wide array of datasets when optimizing over featurization and hyperparameter decisions.}
}