GPflowOpt: A Bayesian Optimization Library using TensorFlow

TL;DR

GPflowOpt is a flexible, scalable Python library for Bayesian optimization built on TensorFlow, enabling automatic differentiation, GPU acceleration, and easy customization of models and acquisition functions.

Contribution

It introduces a new Bayesian optimization framework leveraging GPflow and TensorFlow, with support for custom models, acquisition functions, and multi-objective optimization.

Findings

Includes standard and advanced acquisition functions

Supports multi-objective Bayesian optimization

Offers scalability and customization features

Abstract

A novel Python framework for Bayesian optimization known as GPflowOpt is introduced. The package is based on the popular GPflow library for Gaussian processes, leveraging the benefits of TensorFlow including automatic differentiation, parallelization and GPU computations for Bayesian optimization. Design goals focus on a framework that is easy to extend with custom acquisition functions and models. The framework is thoroughly tested and well documented, and provides scalability. The current released version of GPflowOpt includes some standard single-objective acquisition functions, the state-of-the-art max-value entropy search, as well as a Bayesian multi-objective approach. Finally, it permits easy use of custom modeling strategies implemented in GPflow.