Bayesian optimization with Gaussian processes: Insights from hyperspectral trait search
Azam, Ruhana
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/122148
Description
Title
Bayesian optimization with Gaussian processes: Insights from hyperspectral trait search
Author(s)
Azam, Ruhana
Issue Date
2023-12-04
Director of Research (if dissertation) or Advisor (if thesis)
Koyejo, Sanmi
Department of Study
Computer Science
Discipline
Computer Science
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
Bayesian Optimization
Gaussian Process
Genomic prediction
Abstract
The application of Bayesian Optimization using Gaussian Processes (BO-GP) for global optimization problems is ubiquitous across scientific disciplines because, beyond good performance, it supports exact inference, is interpretable, and has straightforward uncertainty quantification. In this thesis, we reexamine the biological application of BO-GP in searching trait spaces for genomic prediction, which uses genome-wide marker information to predict breeding values for agronomically important traits. Genomic predictions help breeders select desirable plants earlier in the field season without waiting to observe traits later. To reduce costs of collecting data for genomic prediction models, low-cost, hyperspectral data, which are highly correlated with desired traits, can be utilized as a proxy. While these hyperspectral spaces are known to be sharp and aperiodic, BO-GP is considered a feasible approach. However, our work finds that a simple random search surprisingly achieves comparable performance to BO-GP while requiring significantly less computing cost. Through a careful investigation, we can explain this observation as a limitation of the standard implementation and use of BO-GP (e.g., the default use of Matérn kernels), for sharp and aperiodic functions -- where the incompatible structure results in samples similar to random search but with higher computational cost.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
