Practical Path-based Bayesian Optimization
Abstract
There has been a surge in interest in data-driven experimental design with applications to chemical engineering and drug manufacturing. Bayesian optimization (BO) has proven to be adaptable to such cases, since we can model the reactions of interest as expensive black-box functions. Sometimes, the cost of this black-box functions can be separated into two parts: (a) the cost of the experiment itself, and (b) the cost of changing the input parameters. In this short paper, we extend the SnAKe algorithm to deal with both types of costs simultaneously. We further propose extensions to the case of a maximum allowable input change, as well as to the multi-objective setting.
Keywords
Cite
@article{arxiv.2312.00622,
title = {Practical Path-based Bayesian Optimization},
author = {Jose Pablo Folch and James Odgers and Shiqiang Zhang and Robert M Lee and Behrang Shafei and David Walz and Calvin Tsay and Mark van der Wilk and Ruth Misener},
journal= {arXiv preprint arXiv:2312.00622},
year = {2023}
}
Comments
6 main pages, 12 with references and appendix. 4 figures, 2 tables. To appear in NeurIPS 2023 Workshop on Adaptive Experimental Design and Active Learning in the Real World