Local Acquisition Function for Active Level Set Estimation
Abstract
In this paper, we propose a new acquisition function based on local search for active super-level set estimation. Conventional acquisition functions for level set estimation problems are considered to struggle with problems where the threshold is high, and many points in the upper-level set have function values close to the threshold. The proposed method addresses this issue by effectively switching between two acquisition functions: one rapidly finds local level set and the other performs global exploration. The effectiveness of the proposed method is evaluated through experiments with synthetic and real-world datasets.
Cite
Text
Kokubun et al. "Local Acquisition Function for Active Level Set Estimation." NeurIPS 2023 Workshops: ReALML, 2023.Markdown
[Kokubun et al. "Local Acquisition Function for Active Level Set Estimation." NeurIPS 2023 Workshops: ReALML, 2023.](https://mlanthology.org/neuripsw/2023/kokubun2023neuripsw-local/)BibTeX
@inproceedings{kokubun2023neuripsw-local,
title = {{Local Acquisition Function for Active Level Set Estimation}},
author = {Kokubun, Yuta and Matsui, Kota and Kutsukake, Kentaro and Kumagai, Wataru and Kanamori, Takafumi},
booktitle = {NeurIPS 2023 Workshops: ReALML},
year = {2023},
url = {https://mlanthology.org/neuripsw/2023/kokubun2023neuripsw-local/}
}