Preprint · 2022 · Open access

PI is back! Switching Acquisition Functions in Bayesian Optimization

Carolin Benjamins, Elena Raponi, Anja Jankovič, Koen van der Blom, Maria Laura Santoni, Marius Lindauer, Carola Doerr

arXiv (Cornell University) · Published 2 November 2022 · DOI 10.48550/arxiv.2211.01455

Open PDFView at the publisherCite this paper

In brief

“PI is back! Switching Acquisition Functions in Bayesian Optimization” is a 2022 preprint by Carolin Benjamins and 6 co-authors, published in arXiv (Cornell University). OpenAlex counts 18 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
18
Most cited in
2025 (10)
References
8
Authors
7

Abstract

Bayesian Optimization (BO) is a powerful, sample-efficient technique to optimize expensive-to-evaluate functions. Each of the BO components, such as the surrogate model, the acquisition function (AF), or the initial design, is subject to a wide range of design choices. Selecting the right components for a given optimization task is a challenging task, which can have significant impact on the quality of the obtained results. In this work, we initiate the analysis of which AF to favor for which optimization scenarios. To this end, we benchmark SMAC3 using Expected Improvement (EI) and Probability of Improvement (PI) as acquisition functions on the 24 BBOB functions of the COCO environment. We compare their results with those of schedules switching between AFs. One schedule aims to use EI's explorative behavior in the early optimization steps, and then switches to PI for a better exploitation in the final steps. We also compare this to a random schedule and round-robin selection of EI and PI. We observe that dynamic schedules oftentimes outperform any single static one. Our results suggest that a schedule that allocates the first 25 % of the optimization budget to EI and the last 75 % to PI is a reliable default. However, we also observe considerable performance differences for the 24 functions, suggesting that a per-instance allocation, possibly learned on the fly, could offer significant improvement over the state-of-the-art BO designs.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Benjamins, C., Raponi, E., Jankovič, A., van der Blom, K., Santoni, M. L., Lindauer, M., & Doerr, C. (2022). PI is back! Switching Acquisition Functions in Bayesian Optimization. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2211.01455

MLA 9

Benjamins, Carolin, et al. “PI is back! Switching Acquisition Functions in Bayesian Optimization.” arXiv (Cornell University), 2022. https://doi.org/10.48550/arxiv.2211.01455.

Chicago

Benjamins, Carolin, Elena Raponi, Anja Jankovič, Koen van der Blom, Maria Laura Santoni, Marius Lindauer, and Carola Doerr. 2022. “PI is back! Switching Acquisition Functions in Bayesian Optimization.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2211.01455.

Harvard

Benjamins, C. et al. (2022) ‘PI is back! Switching Acquisition Functions in Bayesian Optimization’, arXiv (Cornell University). doi:10.48550/arxiv.2211.01455.

IEEE

C. Benjamins et al., “PI is back! Switching Acquisition Functions in Bayesian Optimization,” arXiv (Cornell University), 2022, doi: 10.48550/arxiv.2211.01455.

BibTeX

@misc{benjamins2022back,
  title     = {{PI is back! Switching Acquisition Functions in Bayesian Optimization}},
  author    = {Benjamins, Carolin and Raponi, Elena and Jankovič, Anja and van der Blom, Koen and Santoni, Maria Laura and Lindauer, Marius and Doerr, Carola},
  howpublished = {arXiv (Cornell University)},
  year      = {2022},
  doi       = {10.48550/arxiv.2211.01455},
  url       = {https://doi.org/10.48550/arxiv.2211.01455}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202333
20251010
202655

Topic and keywords

Social Sciences › Decision Sciences › Management Science and Operations Research › Advanced Bandit Algorithms Research

  • Bayesian optimization
  • acquisition functions
  • probability of improvement
  • expected improvement
  • BBOB benchmark
  • COCO
  • expensive-to-evaluate functions
  • exploration-exploitation balance

Work with this paper

Most-cited papers in Decision Sciences

All highly cited Decision Sciences papers

Questions about this paper

How do I cite this paper in APA style?

Benjamins, C., Raponi, E., Jankovič, A., van der Blom, K., Santoni, M. L., Lindauer, M., & Doerr, C. (2022). PI is back! Switching Acquisition Functions in Bayesian Optimization. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2211.01455

How many times has this paper been cited?

18 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Carolin Benjamins, Elena Raponi, Anja Jankovič, Koen van der Blom, Maria Laura Santoni, Marius Lindauer, Carola Doerr.

What is the DOI of this paper?

10.48550/arxiv.2211.01455, permanently at https://doi.org/10.48550/arxiv.2211.01455. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.