Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisation

Abstract

In order to improve the performance of Bayesian optimisation, we develop a modified Gaussian process upper confidence bound (GP-UCB) acquisition function. This is done by sampling the exploration-exploitation trade-off parameter from a distribution. We prove that this allows the expected trade-off parameter to be altered to better suit the problem without compromising a bound on the function's Bayesian regret. We also provide results showing that our method achieves better performance than GP-UCB in a range of real-world and synthetic problems.

Cite

Text

Berk et al. "Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisation." International Joint Conference on Artificial Intelligence, 2020. doi:10.24963/IJCAI.2020/316

Markdown

[Berk et al. "Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisation." International Joint Conference on Artificial Intelligence, 2020.](https://mlanthology.org/ijcai/2020/berk2020ijcai-randomised/) doi:10.24963/IJCAI.2020/316

BibTeX

@inproceedings{berk2020ijcai-randomised,
  title     = {{Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisation}},
  author    = {Berk, Julian and Gupta, Sunil and Rana, Santu and Venkatesh, Svetha},
  booktitle = {International Joint Conference on Artificial Intelligence},
  year      = {2020},
  pages     = {2284-2290},
  doi       = {10.24963/IJCAI.2020/316},
  url       = {https://mlanthology.org/ijcai/2020/berk2020ijcai-randomised/}
}