Advances in Bayesian Optimization with Applications in Aerospace Engineering
Optimization requires the quantities of interest that define objective functions and constraints to be evaluated a large number of times. In aerospace engineering, these quantities of interest can be expensive to compute (e.g., numerically solving a set of partial differential equations), leading to...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
American Institute of Aeronautics and Astronautics
2018
|
| Online Access: | http://hdl.handle.net/1721.1/116471 https://orcid.org/0000-0003-4222-5358 https://orcid.org/0000-0003-2156-9338 |
| _version_ | 1826202512997220352 |
|---|---|
| author | Poloczek, Matthias Frazier, Peter Lam, Remi Roger Alain Paul Willcox, Karen E |
| author2 | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
| author_facet | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Poloczek, Matthias Frazier, Peter Lam, Remi Roger Alain Paul Willcox, Karen E |
| author_sort | Poloczek, Matthias |
| collection | MIT |
| description | Optimization requires the quantities of interest that define objective functions and constraints to be evaluated a large number of times. In aerospace engineering, these quantities of interest can be expensive to compute (e.g., numerically solving a set of partial differential equations), leading to a challenging optimization problem. Bayesian optimization (BO) is a
class of algorithms for the global optimization of expensive-to-evaluate functions. BO leverages all past evaluations available to construct a surrogate model. This surrogate model is then used to select the next design to evaluate. This paper reviews two recent advances in BO that tackle the challenges of optimizing expensive functions and thus can enrich the
optimization toolbox of the aerospace engineer. The first method addresses optimization problems subject to inequality constraints where a finite budget of evaluations is available, a common situation when dealing with expensive models (e.g., a limited time to conduct the optimization study or limited access to a supercomputer). This challenge is addressed via a lookahead BO algorithm that plans the sequence of designs to evaluate in order to maximize the improvement achieved, not only at the next iteration, but once the total budget is consumed. The second method demonstrates how sensitivity information, such as gradients computed with adjoint methods, can be incorporated into a BO algorithm. This algorithm exploits sensitivity information in two ways: first, to enhance the surrogate model, and second, to improve the selection of the next design to evaluate by accounting for future gradient evaluations. The benefits of the two methods are demonstrated on aerospace examples. |
| first_indexed | 2024-09-23T12:08:41Z |
| format | Article |
| id | mit-1721.1/116471 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T12:08:41Z |
| publishDate | 2018 |
| publisher | American Institute of Aeronautics and Astronautics |
| record_format | dspace |
| spelling | mit-1721.1/1164712022-09-28T00:28:31Z Advances in Bayesian Optimization with Applications in Aerospace Engineering Poloczek, Matthias Frazier, Peter Lam, Remi Roger Alain Paul Willcox, Karen E Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Willcox, Karen E Lam, Remi Willcox, Karen E Optimization requires the quantities of interest that define objective functions and constraints to be evaluated a large number of times. In aerospace engineering, these quantities of interest can be expensive to compute (e.g., numerically solving a set of partial differential equations), leading to a challenging optimization problem. Bayesian optimization (BO) is a class of algorithms for the global optimization of expensive-to-evaluate functions. BO leverages all past evaluations available to construct a surrogate model. This surrogate model is then used to select the next design to evaluate. This paper reviews two recent advances in BO that tackle the challenges of optimizing expensive functions and thus can enrich the optimization toolbox of the aerospace engineer. The first method addresses optimization problems subject to inequality constraints where a finite budget of evaluations is available, a common situation when dealing with expensive models (e.g., a limited time to conduct the optimization study or limited access to a supercomputer). This challenge is addressed via a lookahead BO algorithm that plans the sequence of designs to evaluate in order to maximize the improvement achieved, not only at the next iteration, but once the total budget is consumed. The second method demonstrates how sensitivity information, such as gradients computed with adjoint methods, can be incorporated into a BO algorithm. This algorithm exploits sensitivity information in two ways: first, to enhance the surrogate model, and second, to improve the selection of the next design to evaluate by accounting for future gradient evaluations. The benefits of the two methods are demonstrated on aerospace examples. 2018-06-21T14:37:16Z 2018-06-21T14:37:16Z 2018-01 Article http://purl.org/eprint/type/ConferencePaper 978-1-62410-529-6 http://hdl.handle.net/1721.1/116471 Lam, Rémi, et al. "Advances in Bayesian Optimization with Applications in Aerospace Engineering." 2018 AIAA Non-Deterministic Approaches Conference, 8-12 January, 2018, Kissimmee, Florida, American Institute of Aeronautics and Astronautics, 2018. https://orcid.org/0000-0003-4222-5358 https://orcid.org/0000-0003-2156-9338 en_US https://doi.org/10.2514/6.2018-1656 2018 AIAA Non-Deterministic Approaches Conference Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf American Institute of Aeronautics and Astronautics Prof. Willcox via Barbara Williams |
| spellingShingle | Poloczek, Matthias Frazier, Peter Lam, Remi Roger Alain Paul Willcox, Karen E Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title | Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title_full | Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title_fullStr | Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title_full_unstemmed | Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title_short | Advances in Bayesian Optimization with Applications in Aerospace Engineering |
| title_sort | advances in bayesian optimization with applications in aerospace engineering |
| url | http://hdl.handle.net/1721.1/116471 https://orcid.org/0000-0003-4222-5358 https://orcid.org/0000-0003-2156-9338 |
| work_keys_str_mv | AT poloczekmatthias advancesinbayesianoptimizationwithapplicationsinaerospaceengineering AT frazierpeter advancesinbayesianoptimizationwithapplicationsinaerospaceengineering AT lamremirogeralainpaul advancesinbayesianoptimizationwithapplicationsinaerospaceengineering AT willcoxkarene advancesinbayesianoptimizationwithapplicationsinaerospaceengineering |