Keywords: Bayesian optimization, global optimization, Gaussian process, combinatorial optimization, high-dimensional
TL;DR: We propose a Bayesian optimization algorithm for combinatorial, mixed, and continuous spaces that is robust and gives state-of-the-art performance on a wide set of benchmarks.
Abstract: Impactful applications such as materials discovery, hardware design, neural architecture search, or portfolio optimization require optimizing high-dimensional black-box functions with mixed and combinatorial input spaces.
While Bayesian optimization has recently made significant progress in solving such problems, an in-depth analysis reveals that the current state-of-the-art methods are not reliable.
Their performances degrade substantially when the unknown optima of the function do not have a certain structure.
To fill the need for a reliable algorithm for combinatorial and mixed spaces, this paper proposes Bounce that relies on a novel map of various variable types into nested embeddings of increasing dimensionality.
Comprehensive experiments show that Bounce reliably achieves and often even improves upon state-of-the-art performance on a variety of high-dimensional problems.
Supplementary Material: zip
Submission Number: 6886
Loading