A Weight Inheritance and Guidance Strategy-Based Evolutionary Network Architecture Search
IEEE Transactions on Evolutionary Computation · 2026-04 · 30 (2) · 549-563
DOI: 10.1109/tevc.2025.3558334 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Evolutionary Computation, 2026-04. Verfasst von Li Yan und 5 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 66 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Innovation Team of Colleges and Universities in Henan Province, Science and Technology Innovation Talents of Colleges and Universities in Henan Province.
Abstract (Originalsprache)
Neural architecture search (NAS) has emerged as an important area in deep learning since it can automatically design high-performance network architectures, where evolution-based NAS (EvoNAS) has made great progress due to the efficient optimization ability of evolutionary algorithms. However, EvoNAS requires evaluating the architectures formed by individuals in the population, and it is inevitable to consume a large amount of evaluation time and computational resources, resulting in restricting the applicability of EvoNAS. To solve the above problems, this article proposes a weight inheritance and guided strategy-based evolutionary network architecture search (WIGEvoNAS). First, based on existing manually designed networks, an expanded search space is designed, which includes new convolution operations. Second, a weight inheritance strategy is proposed to reduce the training time of candidate architectures in each generation. Finally, a guidance mutation strategy is proposed to direct population evolution toward architectures with superior for the purpose of generating better offspring. The proposed method is compared with several state-of-the-art NAS methods and manual networks on the CIFAR-10, CIFAR-100, and the NASBench-201 benchmark datasets. The empirical results demonstrate that the proposed method achieves promising performance, with error rates of 2.52% on CIFAR-10 and 15.43% on CIFAR-100 respectively. Moreover, the proposed method significantly reduces search costs to 0.9 GPU-days.
Förderung
- National Natural Science Foundation of China
- Science and Technology Innovation Team of Colleges and Universities in Henan Province
- Science and Technology Innovation Talents of Colleges and Universities in Henan Province
- Key Science and Technology Projects of Henan Province
- Research Award Fund for Outstanding Young Teachers in Henan Provincial Institutions of Higher Education
- Key Research and Development Projects of Henan Province
- Graduate Education Reform Project of Henan Province
- International Science and Technology Cooperation Program in Henan Province
Zitieren
APA
Yan, L., Du, Y., Liang, J., Qu, B., Li, C., & Yu, K. (2026). A Weight Inheritance and Guidance Strategy-Based Evolutionary Network Architecture Search. IEEE Transactions on Evolutionary Computation, 30(2), 549-563. https://doi.org/10.1109/tevc.2025.3558334
Vancouver
Yan L, Du Y, Liang J, Qu B, Li C, Yu K. A Weight Inheritance and Guidance Strategy-Based Evolutionary Network Architecture Search. IEEE Transactions on Evolutionary Computation. 2026;30(2):549-563. doi:10.1109/tevc.2025.3558334
BibTeX
@article{Yan2026,
author = {Yan, Li and Du, Yi and Liang, Jing and Qu, Boyang and Li, Chao and Yu, Kunjie},
title = {A Weight Inheritance and Guidance Strategy-Based Evolutionary Network Architecture Search},
journal = {IEEE Transactions on Evolutionary Computation},
year = {2026},
volume = {30},
number = {2},
pages = {549--563},
doi = {10.1109/tevc.2025.3558334}
}
EUVIMED