Paper · 2020 · Open access

High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces

Riccardo Moriconi

Published 1 November 2020 · DOI 10.25560/91394

Open full textView at the publisherCite this paper

In brief

“High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces” is a 2020 paper by Riccardo Moriconi. OpenAlex counts 0 citations as of 19 November 2026. It is open access under a CC BY-NC licence: the full text is free to read at Spiral (Imperial College London).

Citations
0
References
74
Authors
1

Abstract

Bayesian optimization is a powerful technique for the optimization of expensive black-box functions. It is used in a wide range of applications such as in drug and material design and training of machine learning models, e.g. large deep networks. We propose to extend this approach to high-dimensional settings, that is where the number of parameters to be optimized exceeds 10--20. In this thesis, we scale Bayesian optimization by exploiting different types of projections and the intrinsic low-dimensionality assumption of the objective function. We reformulate the problem in a low-dimensional subspace and learn a response surface and maximize an acquisition function in this low-dimensional projection. Contributions include i) a probabilistic model for axis-aligned projections, such as the quantile-Gaussian process and ii) a probabilistic model for learning a feature space by means of manifold Gaussian processes. In the latter contribution, we propose to learn a low-dimensional feature space jointly with (a) the response surface and (b) a reconstruction mapping. Finally, we present empirical results against well-known baselines in high-dimensional Bayesian optimization and provide possible directions for future research in this field.

Cite this paper

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

APA 7

Moriconi, R. (2020). High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces. https://doi.org/10.25560/91394

MLA 9

Moriconi, Riccardo. “High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces.” 2020. https://doi.org/10.25560/91394.

Chicago

Moriconi, Riccardo. 2020. “High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces.” https://doi.org/10.25560/91394.

Harvard

Moriconi, R. (2020) ‘High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces’. doi:10.25560/91394.

IEEE

R. Moriconi, “High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces,” 2020, doi: 10.25560/91394.

BibTeX

@article{moriconi2020high,
  title     = {{High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces}},
  author    = {Moriconi, Riccardo},
  year      = {2020},
  doi       = {10.25560/91394},
  url       = {https://doi.org/10.25560/91394}
}

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

Topic and keywords

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

  • feature space learning
  • black-box optimization
  • subspace projection

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?

Moriconi, R. (2020). High-dimensional Bayesian optimization with intrinsically low-dimensional response surfaces. https://doi.org/10.25560/91394

How many times has this paper been cited?

0 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 Spiral (Imperial College London): sign in and open the full text from this page.

Who wrote this paper?

Riccardo Moriconi.

What is the DOI of this paper?

10.25560/91394, permanently at https://doi.org/10.25560/91394. 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.