arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.04815cs.LG
收藏

ChaCha用于在线AutoML

ChaCha for Online AutoML

Microsoft Research(微软研究院)

Qingyun Wu, Chi Wang, John Langford, Paul Mineiro, Marco Rossi

摘要:

我们提出了ChaCha(Chaon-Challenge)算法,用于在在线学习设置中在线选择超参数。ChaCha处理确定冠军的过程,并根据样本的复杂性范围随着时间的推移安排一组“活”挑战者。通过基于冠军的应用依赖预言机将最优配置加入考虑范围,保证了算法具有次线性遗憾。从经验上讲,我们表明,ChaCha在优化特征化和超参数决策时,在广泛的数据集上提供了良好的性能。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑