Batch Bayesian optimization
Name
1220836868-MIT.pdf
Size
5.04 MB
Format
Adobe PDF
Checksum (MD5)
c5e44658f95d06ba828e9e31dc8c6e68
Author(s)
Hunt, Nathan(Nathan R.)
Advisor(s)
David K. Gifford.
Date Issued
2020
Publisher
Massachusetts Institute of Technology
Abstract
Bayesian optimization is a useful technique for maximizing expensive, unknown functions that employs an acquisition function to determine what unseen input point to query next. In many real-world applications, batches of input points can be queried simultaneously for only a small marginal cost compared to querying a single point. Most classical acquisition functions cannot be used for batch acquisition, and thus batch acquisition strategies are required. Several such strategies have been developed in the past decade. We review and compare batch acquisition strategies in a variety of settings to assist practitioners in selecting appropriate batch acquisition functions and facilitate further research in this area.
Description
Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, February, 2020
Cataloged from PDF version of thesis. "Pages contain copy print steak marks"--Disclaimer page.
Includes bibliographical references (pages 73-77).
Subjects
Electrical Engineering and Computer Science.
MIT Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
MIT theses may be protected by copyright. Please reuse MIT thesis content according to the MIT Libraries Permissions Policy, which is available through the URL provided.
Persistent DSpace Link
