Multi-Target Optimisation via Bayesian Optimisation and Linear Programming
Abstract
In Bayesian Multi-Objective optimisation, expected hypervolume improvement is often used to measure the goodness of candidate solutions. However when there are many objectives the calculation of expected hypervolume improvement can become computationally prohibitive. An alternative approach measures the goodness of a candidate based on the distance of that candidate from the Pareto front in objective space. In this paper we present a novel distance-based Bayesian Many-Objective optimisation algorithm. We demonstrate the efficacy of our algorithm on three problems, namely the DTLZ2 benchmark problem, a hyper-parameter selection problem, and high-temperature creep-resistant alloy design.
Cite
Text
Shilton et al. "Multi-Target Optimisation via Bayesian Optimisation and Linear Programming." Conference on Uncertainty in Artificial Intelligence, 2018.Markdown
[Shilton et al. "Multi-Target Optimisation via Bayesian Optimisation and Linear Programming." Conference on Uncertainty in Artificial Intelligence, 2018.](https://mlanthology.org/uai/2018/shilton2018uai-multi/)BibTeX
@inproceedings{shilton2018uai-multi,
title = {{Multi-Target Optimisation via Bayesian Optimisation and Linear Programming}},
author = {Shilton, Alistair and Rana, Santu and Gupta, Sunil and Venkatesh, Svetha},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2018},
pages = {145-155},
url = {https://mlanthology.org/uai/2018/shilton2018uai-multi/}
}