Grey-Box Bayesian Optimization: Improving Performance by Looking Inside the Black-Box

dc.contributor.authorToscano Palmerin, Saul
dc.contributor.chairFrazier, Peter
dc.contributor.committeeMemberHenderson, Shane
dc.contributor.committeeMemberBindel, David
dc.date.accessioned2020-08-10T20:24:49Z
dc.date.available2020-08-10T20:24:49Z
dc.date.issued2020-05
dc.description184 pages
dc.description.abstractNon-convex time-consuming objectives are often optimized using “black-box” optimization. These approaches assume very little about the objective. While broadly applicable, these approaches typically require more evaluations than methods exploiting more problem structure. In particular, often, we can acquire information about the objective function in ways other than direct evaluation, which is less time-consuming than evaluating the objective directly. This allows us to develop novel Bayesian optimization algorithms that outperform methods that rely only objective function evaluations. In this thesis, we consider three problems: optimization of sum and integrals of expensive-to-evaluate integrands; optimizing hyperparameters for iteratively trained supervised learning machine learning algorithms; and optimizing non-convex functions with a new efficient multistart stochastic gradient descent algorithm.
dc.identifier.doihttps://doi.org/10.7298/3f24-n006
dc.identifier.otherToscanoPalmerin_cornellgrad_0058F_11869
dc.identifier.otherhttp://dissertations.umi.com/cornellgrad:11869
dc.identifier.urihttps://hdl.handle.net/1813/70478
dc.language.isoen
dc.subjectBayesian optimization
dc.subjectblack-box optimization
dc.subjectGaussian process
dc.titleGrey-Box Bayesian Optimization: Improving Performance by Looking Inside the Black-Box
dc.typedissertation or thesis
dcterms.licensehttps://hdl.handle.net/1813/59810
thesis.degree.disciplineOperations Research and Information Engineering
thesis.degree.grantorCornell University
thesis.degree.levelDoctor of Philosophy
thesis.degree.namePh. D., Operations Research and Information Engineering
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
ToscanoPalmerin_cornellgrad_0058F_11869.pdf
Size:
2.7 MB
Format:
Adobe Portable Document Format