Combining global and local surrogate models to accelerate evolutionary optimization
In this paper, we present a novel surrogate-assisted evolutionary optimization framework for solving computationally expensive problems. The proposed framework uses computationally cheap hierarchical surrogate models constructed through online learning to replace the exact computationally expensive...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2021
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/147970 |
| _version_ | 1811691135701090304 |
|---|---|
| author | Zhou, Zongzhao Ong, Yew-Soon Nair, Prasanth B. Keane, Andy J. Lum, Kai Yew |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Zhou, Zongzhao Ong, Yew-Soon Nair, Prasanth B. Keane, Andy J. Lum, Kai Yew |
| author_sort | Zhou, Zongzhao |
| collection | NTU |
| description | In this paper, we present a novel surrogate-assisted evolutionary optimization framework for solving computationally expensive problems. The proposed framework uses computationally cheap hierarchical surrogate models constructed through online learning to replace the exact computationally expensive objective functions during evolutionary search. At the first level, the framework employs a data-parallel Gaussian process based global surrogate model to filter the evolutionary algorithm (EA) population of promising individuals. Subsequently, these potential individuals undergo a memetic search in the form of Lamarckian learning at the second level. The Lamarckian evolution involves a trust-region enabled gradient-based search strategy that employs radial basis function local surrogate models to accelerate convergence. Numerical results are presented on a series of benchmark test functions and on an aerodynamic shape design problem. The results obtained suggest that the proposed optimization framework converges to good designs on a limited computational budget. Furthermore, it is shown that the new algorithm gives significant savings in computational cost when compared to the traditional evolutionary algorithm and other surrogate assisted optimization frameworks. |
| first_indexed | 2024-10-01T06:15:05Z |
| format | Journal Article |
| id | ntu-10356/147970 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:15:05Z |
| publishDate | 2021 |
| record_format | dspace |
| spelling | ntu-10356/1479702021-04-16T01:37:31Z Combining global and local surrogate models to accelerate evolutionary optimization Zhou, Zongzhao Ong, Yew-Soon Nair, Prasanth B. Keane, Andy J. Lum, Kai Yew School of Computer Science and Engineering Engineering::Computer science and engineering Aerodynamic Shape Design Evolutionary Optimization In this paper, we present a novel surrogate-assisted evolutionary optimization framework for solving computationally expensive problems. The proposed framework uses computationally cheap hierarchical surrogate models constructed through online learning to replace the exact computationally expensive objective functions during evolutionary search. At the first level, the framework employs a data-parallel Gaussian process based global surrogate model to filter the evolutionary algorithm (EA) population of promising individuals. Subsequently, these potential individuals undergo a memetic search in the form of Lamarckian learning at the second level. The Lamarckian evolution involves a trust-region enabled gradient-based search strategy that employs radial basis function local surrogate models to accelerate convergence. Numerical results are presented on a series of benchmark test functions and on an aerodynamic shape design problem. The results obtained suggest that the proposed optimization framework converges to good designs on a limited computational budget. Furthermore, it is shown that the new algorithm gives significant savings in computational cost when compared to the traditional evolutionary algorithm and other surrogate assisted optimization frameworks. Nanyang Technological University Accepted version This work was supported by NTU/SCE under Grant CE-SUG 3/03. 2021-04-16T01:37:30Z 2021-04-16T01:37:30Z 2007 Journal Article Zhou, Z., Ong, Y., Nair, P. B., Keane, A. J. & Lum, K. Y. (2007). Combining global and local surrogate models to accelerate evolutionary optimization. IEEE Transactions On Systems, Man and Cybernetics Part C: Applications and Reviews, 37(1), 66-76. https://dx.doi.org/10.1109/TSMCC.2005.855506 1094-6977 https://hdl.handle.net/10356/147970 10.1109/TSMCC.2005.855506 2-s2.0-33845749639 1 37 66 76 en CE-SUG 3/03 IEEE Transactions on Systems, Man and Cybernetics Part C: Applications and Reviews © 2007 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/TSMCC.2005.855506. application/pdf |
| spellingShingle | Engineering::Computer science and engineering Aerodynamic Shape Design Evolutionary Optimization Zhou, Zongzhao Ong, Yew-Soon Nair, Prasanth B. Keane, Andy J. Lum, Kai Yew Combining global and local surrogate models to accelerate evolutionary optimization |
| title | Combining global and local surrogate models to accelerate evolutionary optimization |
| title_full | Combining global and local surrogate models to accelerate evolutionary optimization |
| title_fullStr | Combining global and local surrogate models to accelerate evolutionary optimization |
| title_full_unstemmed | Combining global and local surrogate models to accelerate evolutionary optimization |
| title_short | Combining global and local surrogate models to accelerate evolutionary optimization |
| title_sort | combining global and local surrogate models to accelerate evolutionary optimization |
| topic | Engineering::Computer science and engineering Aerodynamic Shape Design Evolutionary Optimization |
| url | https://hdl.handle.net/10356/147970 |
| work_keys_str_mv | AT zhouzongzhao combiningglobalandlocalsurrogatemodelstoaccelerateevolutionaryoptimization AT ongyewsoon combiningglobalandlocalsurrogatemodelstoaccelerateevolutionaryoptimization AT nairprasanthb combiningglobalandlocalsurrogatemodelstoaccelerateevolutionaryoptimization AT keaneandyj combiningglobalandlocalsurrogatemodelstoaccelerateevolutionaryoptimization AT lumkaiyew combiningglobalandlocalsurrogatemodelstoaccelerateevolutionaryoptimization |