Bayesian optimization~(BO) is often used for accelerator tuning due to its high sample efficiency. However, the computational scalability of training over large data-set can be problematic and the adoption of historical data in a computationally efficient way is not trivial. Here, we exploit a neural network model trained over historical data as a prior mean of BO for FRIB Front-End tuning.
Cite
@article{arxiv.2211.06400,
title = {Prior-mean-assisted Bayesian optimization application on FRIB Front-End tunning},
author = {Kilean Hwang and Tomofumi Maruta and Alexander Plastun and Kei Fukushima and Tong Zhang and Qiang Zhao and Peter Ostroumov and Yue Hao},
journal= {arXiv preprint arXiv:2211.06400},
year = {2022}
}