Preprint · 2020 · Open access

Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization

Benjamin Letham, Roberto Calandra, Akshara Rai, Eytan Bakshy

arXiv (Cornell University), 33, 1546–1558 · Published 31 January 2020 · DOI 10.48550/arxiv.2001.11659

Open PDFView at the publisherCite this paper

In brief

“Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization” is a 2020 preprint by Benjamin Letham and 3 co-authors, published in arXiv (Cornell University). OpenAlex counts 159 citations as of 19 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
159
Most cited in
2024 (33)
References
56
Authors
4

Abstract

Bayesian optimization (BO) is a popular approach to optimize expensive-to-evaluate black-box functions. A significant challenge in BO is to scale to high-dimensional parameter spaces while retaining sample efficiency. A solution considered in existing literature is to embed the high-dimensional space in a lower-dimensional manifold, often via a random linear embedding. In this paper, we identify several crucial issues and misconceptions about the use of linear embeddings for BO. We study the properties of linear embeddings from the literature and show that some of the design choices in current approaches adversely impact their performance. We show empirically that properly addressing these issues significantly improves the efficacy of linear embeddings for BO on a range of problems, including learning a gait policy for robot locomotion.

Cite this paper

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

APA 7

Letham, B., Calandra, R., Rai, A., & Bakshy, E. (2020). Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization. arXiv (Cornell University), 33, 1546–1558. https://doi.org/10.48550/arxiv.2001.11659

MLA 9

Letham, Benjamin, et al. “Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization.” arXiv (Cornell University), vol. 33, 2020, pp. 1546–1558. https://doi.org/10.48550/arxiv.2001.11659.

Chicago

Letham, Benjamin, Roberto Calandra, Akshara Rai, and Eytan Bakshy. 2020. “Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization.” arXiv (Cornell University) 33: 1546–1558. https://doi.org/10.48550/arxiv.2001.11659.

Harvard

Letham, B. et al. (2020) ‘Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization’, arXiv (Cornell University), 33, pp. 1546–1558. doi:10.48550/arxiv.2001.11659.

IEEE

B. Letham, R. Calandra, A. Rai, and E. Bakshy, “Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization,” arXiv (Cornell University), vol. 33, pp. 1546–1558, 2020, doi: 10.48550/arxiv.2001.11659.

BibTeX

@misc{letham2020examining,
  title     = {{Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization}},
  author    = {Letham, Benjamin and Calandra, Roberto and Rai, Akshara and Bakshy, Eytan},
  howpublished = {arXiv (Cornell University)},
  year      = {2020},
  volume    = {33},
  pages     = {1546--1558},
  doi       = {10.48550/arxiv.2001.11659},
  url       = {https://doi.org/10.48550/arxiv.2001.11659}
}

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

Citations by year

YearCitationsShare of the best year
201911
202044
20211616
20222222
20232626
20243333
20253232
20262525

Topic and keywords

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

  • Bayesian optimization
  • linear embedding
  • high-dimensional parameter spaces
  • random embeddings
  • black-box optimization
  • sample efficiency
  • robot locomotion

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?

Letham, B., Calandra, R., Rai, A., & Bakshy, E. (2020). Re-Examining Linear Embeddings for High-Dimensional Bayesian Optimization. arXiv (Cornell University), 33, 1546–1558. https://doi.org/10.48550/arxiv.2001.11659

How many times has this paper been cited?

159 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?

Benjamin Letham, Roberto Calandra, Akshara Rai, Eytan Bakshy.

What is the DOI of this paper?

10.48550/arxiv.2001.11659, permanently at https://doi.org/10.48550/arxiv.2001.11659. 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.