Scalable Bayesian Optimization Using Deep Neural Networks

TL;DR

This paper introduces a neural network-based surrogate model for Bayesian optimization that scales linearly with data, enabling highly parallel hyperparameter tuning for large models and complex tasks.

Contribution

It demonstrates that neural networks can replace Gaussian processes in Bayesian optimization, significantly improving scalability and parallelism.

Findings

Neural network surrogates perform competitively with GPs.

Linear scaling enables high degrees of parallelism.

Effective for large-scale hyperparameter optimization.

Abstract

Bayesian optimization is an effective methodology for the global optimization of functions with expensive evaluations. It relies on querying a distribution over functions defined by a relatively cheap surrogate model. An accurate model for this distribution over functions is critical to the effectiveness of the approach, and is typically fit using Gaussian processes (GPs). However, since GPs scale cubically with the number of observations, it has been challenging to handle objectives whose optimization requires many evaluations, and as such, massively parallelizing the optimization. In this work, we explore the use of neural networks as an alternative to GPs to model distributions over functions. We show that performing adaptive basis function regression with a neural network as the parametric form performs competitively with state-of-the-art GP-based approaches, but scales linearly with the number of data rather than cubically. This allows us to achieve a previously intractable degree of parallelism, which we apply to large scale hyperparameter optimization, rapidly finding competitive models on benchmark object recognition tasks using convolutional networks, and image caption generation using neural language models.