Bounce: Reliable High-Dimensional Bayesian Optimization for Combinatorial and Mixed Spaces

arXiv:2307.00618v2 Announce Type: replace
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 …

abstract analysis applications architecture art arxiv bayesian box cs.lg current design discovery functions hardware materials mixed neural architecture search optimization portfolio progress search spaces state type