An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization
Zhang X, Yu G, Jin Y, Qian F (2023)
NEUROCOMPUTING 538: 126212.
Zeitschriftenaufsatz
| Veröffentlicht | Englisch
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Autor*in
Zhang, Xi;
Yu, Guo;
Jin, YaochuUniBi
;
Qian, Feng
![](https://pub.uni-bielefeld.de/images/icon_orcid.png)
Abstract / Bemerkung
Expensive dynamic multi-objective optimization problems (EDMOPs) is one kind of DMOPs where the objectives change over time and the function evaluations commonly involve computationally intensive simulations or costly physical experiments. Hence, the key to solve EDMOPs is to quickly and accurately track the time-varying Pareto optimal fronts under the limit of small number of function evaluations, in which how to augment enough training data to build informative surrogate models and manage the mod-els during the search process. To overcome the issue, we propose a transfer learning based surrogate assisted evolutionary algorithm (TrSA-DMOEA) to efficiently solve EDMOPs. Specifically, when a change occurs, we propose a knee point-based manifold transfer learning method based on geodesic flow kernel, which exploits the knowledge from previous high-quality knee solutions to augment the training data for building Gaussian process models, thereby improving the computational complexity and the quality of solutions. Moreover, to efficiently find the optima with limited budget of function evaluations, a novel surrogate-assisted mechanism based on an adaptive acquisition function is introduced, which achieves a balance between convergence and diversity by adaptively adjusting the weights of the angle -penalized distance and average uncertainty at different search stages. By comparing with state-of-the-art algorithms on widely used test problems, the experimental results demonstrate that the proposed method outperforms others and is able to efficiently solve EDMOPs.(c) 2023 Elsevier B.V. All rights reserved.
Stichworte
Dynamic multi-objective optimization;
Transfer learning;
Gaussian;
process;
Surrogate assisted evolutionary algorithm
Erscheinungsjahr
2023
Zeitschriftentitel
NEUROCOMPUTING
Band
538
Art.-Nr.
126212
ISSN
0925-2312
eISSN
1872-8286
Page URI
https://pub.uni-bielefeld.de/record/2979725
Zitieren
Zhang X, Yu G, Jin Y, Qian F. An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization. NEUROCOMPUTING. 2023;538: 126212.
Zhang, X., Yu, G., Jin, Y., & Qian, F. (2023). An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization. NEUROCOMPUTING, 538, 126212. https://doi.org/10.1016/j.neucom.2023.03.073
Zhang, Xi, Yu, Guo, Jin, Yaochu, and Qian, Feng. 2023. “An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization”. NEUROCOMPUTING 538: 126212.
Zhang, X., Yu, G., Jin, Y., and Qian, F. (2023). An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization. NEUROCOMPUTING 538:126212.
Zhang, X., et al., 2023. An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization. NEUROCOMPUTING, 538: 126212.
X. Zhang, et al., “An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization”, NEUROCOMPUTING, vol. 538, 2023, : 126212.
Zhang, X., Yu, G., Jin, Y., Qian, F.: An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization. NEUROCOMPUTING. 538, : 126212 (2023).
Zhang, Xi, Yu, Guo, Jin, Yaochu, and Qian, Feng. “An adaptive Gaussian process based manifold transfer learning to expensive dynamic multi-objective optimization”. NEUROCOMPUTING 538 (2023): 126212.
Export
Markieren/ Markierung löschen
Markierte Publikationen
Web of Science
Dieser Datensatz im Web of Science®Suchen in