Squirrel: A Switching Hyperparameter Optimizer
This paper addresses the problem of finding an effective hyperparameter optimization strategy for machine learning practitioners, by combining existing methods.
The paper describes Squirrel, a hyperparameter optimizer that switches between different optimizers, and was submitted to the NeurIPS 2020 BBO challenge.
In this short note, we describe our submission to the NeurIPS 2020 BBO challenge. Motivated by the fact that different optimizers work well on different problems, our approach switches between different optimizers. Since the team names on the competition's leaderboard were randomly generated "alliteration nicknames", consisting of an adjective and an animal with the same initial letter, we called our approach the Switching Squirrel, or here, short, Squirrel.