Few-Shot Bayesian Optimization with Deep Kernel Surrogates
This addresses the computational bottleneck in HPO for machine learning practitioners by enabling faster optimization with fewer evaluations, though it is an incremental improvement over existing transfer learning approaches.
The paper tackles hyperparameter optimization (HPO) by reframing it as a few-shot learning problem, using a meta-learned deep kernel network as a Gaussian process surrogate to adapt quickly to new tasks with few evaluations, achieving state-of-the-art results on diverse metadata sets.
Hyperparameter optimization (HPO) is a central pillar in the automation of machine learning solutions and is mainly performed via Bayesian optimization, where a parametric surrogate is learned to approximate the black box response function (e.g. validation error). Unfortunately, evaluating the response function is computationally intensive. As a remedy, earlier work emphasizes the need for transfer learning surrogates which learn to optimize hyperparameters for an algorithm from other tasks. In contrast to previous work, we propose to rethink HPO as a few-shot learning problem in which we train a shared deep surrogate model to quickly adapt (with few response evaluations) to the response function of a new task. We propose the use of a deep kernel network for a Gaussian process surrogate that is meta-learned in an end-to-end fashion in order to jointly approximate the response functions of a collection of training data sets. As a result, the novel few-shot optimization of our deep kernel surrogate leads to new state-of-the-art results at HPO compared to several recent methods on diverse metadata sets.