RubberBand
Ujval Misra, Richard Liaw, Lisa Dunlap, Romil Bhardwaj, Kirthevasan Kandasamy, Joseph E. Gonzalez, Ion Gabriel Stoica, Alexey Tumanov · 2021
Hyperparameter tuning is essential to achieving state-of-the-art accuracy in machine learning (ML), but requires substantial compute resources to perform. Existing systems primarily focus on effectively allocating resources for a hyperparameter tuning job under fixed resource constraints. We show that the available parallelism in such jobs changes dynamically over the course of execution and, therefore, presents an opportunity to leverage the elasticity of the cloud.