Dissertation · 2023 · Open access

A Framework For Bayesian Optimization of Probabilistic Robustness Metrics

Neel Shah

TSpace (University of Toronto) · Published 1 March 2023

Open PDFView at the publisherCite this paper

In brief

“A Framework For Bayesian Optimization of Probabilistic Robustness Metrics” is a 2023 dissertation by Neel Shah, published in TSpace (University of Toronto). OpenAlex counts 0 citations as of 19 November 2026. It is open access: the full text is free to read at TSpace (University of Toronto).

Citations
0
Authors
1

Abstract

In this thesis, we consider the problem of optimizing probabilistic robustness metrics that depend on the outputs of expensive black-box simulators. In the present work, Gaussian process models of the black-box simulators are leveraged as surrogates to construct sample-based approximations of acquisition functions that can then be used to guide Bayesian optimization algorithms towards the optimum of the probabilistic robustness metric. Sample-based approximations enable accelerated optimization of some acquisition functions using stochastic gradient descent methods, while enabling our framework to accommodate arbitrary robustness metrics. An approach is presented to address the challenges associated with visualizing convergence of probabilistic robustness metrics calculated using surrogate models that are updated sequentially with each observation. A prototype GUI is developed to allow users to specify their problem statement while ensuring all parameters and data are correctly defined. Lastly, numerical studies are presented to illustrate the efficacy of the proposed approach.

Cite this paper

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

APA 7

Shah, N. (2023). A Framework For Bayesian Optimization of Probabilistic Robustness Metrics. TSpace (University of Toronto). http://hdl.handle.net/1807/126808

MLA 9

Shah, Neel. “A Framework For Bayesian Optimization of Probabilistic Robustness Metrics.” TSpace (University of Toronto), 2023. http://hdl.handle.net/1807/126808.

Chicago

Shah, Neel. 2023. “A Framework For Bayesian Optimization of Probabilistic Robustness Metrics.” TSpace (University of Toronto). http://hdl.handle.net/1807/126808.

Harvard

Shah, N. (2023) ‘A Framework For Bayesian Optimization of Probabilistic Robustness Metrics’, TSpace (University of Toronto). Available at: http://hdl.handle.net/1807/126808.

IEEE

N. Shah, “A Framework For Bayesian Optimization of Probabilistic Robustness Metrics,” TSpace (University of Toronto), 2023.

BibTeX

@phdthesis{shah2023framework,
  title     = {{A Framework For Bayesian Optimization of Probabilistic Robustness Metrics}},
  author    = {Shah, Neel},
  year      = {2023},
  url       = {http://hdl.handle.net/1807/126808}
}

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

Topic and keywords

Physical Sciences › Computer Science › Computational Theory and Mathematics › Advanced Multi-Objective Optimization Algorithms

  • Bayesian optimization
  • Kriging surrogate model
  • acquisition functions
  • stochastic gradient descent
  • graphical user interface
  • surrogate model updating

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Shah, N. (2023). A Framework For Bayesian Optimization of Probabilistic Robustness Metrics. TSpace (University of Toronto). http://hdl.handle.net/1807/126808

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 TSpace (University of Toronto): sign in and open the PDF from this page.

Who wrote this paper?

Neel Shah.

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.