HyperJump: Accelerating HyperBand via Risk Modelling

TL;DR

HyperJump enhances HyperBand by integrating risk modeling to skip low-risk configurations, significantly accelerating hyper-parameter tuning across various machine learning tasks with over tenfold speed improvements.

Contribution

This paper introduces HyperJump, a novel method that combines HyperBand's strategy with risk modeling to speed up hyper-parameter optimization by skipping unlikely configurations.

Findings

Over one-order of magnitude speed-up in hyper-parameter tuning

Effective in deep learning, kernel methods, and neural architecture search

Outperforms HyperBand and state-of-the-art optimizers

Abstract

In the literature on hyper-parameter tuning, a number of recent solutions rely on low-fidelity observations (e.g., training with sub-sampled datasets) in order to efficiently identify promising configurations to be then tested via high-fidelity observations (e.g., using the full dataset). Among these, HyperBand is arguably one of the most popular solutions, due to its efficiency and theoretically provable robustness. In this work, we introduce HyperJump, a new approach that builds on HyperBand's robust search strategy and complements it with novel model-based risk analysis techniques that accelerate the search by skipping the evaluation of low risk configurations, i.e., configurations that are likely to be eventually discarded by HyperBand. We evaluate HyperJump on a suite of hyper-parameter optimization problems and show that it provides over one-order of magnitude speed-ups, both in sequential and parallel deployments, on a variety of deep-learning, kernel-based learning, and neural architectural search problems when compared to HyperBand and to several state-of-the-art optimizers.