Timezone: »

 
Spotlight
ChaCha for Online AutoML
Qingyun Wu · Chi Wang · John Langford · Paul Mineiro · Marco Rossi

Wed Jul 21 06:40 AM -- 06:45 AM (PDT) @ None

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.

Author Information

Qingyun Wu (Microsoft Research)
Chi Wang (Microsoft Research)
John Langford (Microsoft Research)
Paul Mineiro (Microsoft)
Marco Rossi (Microsoft Corporation)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors