Meta-Guided Domain-Aware Neural Architecture Search for Rapid Cross- Domain Adaptation
DOI:
https://doi.org/10.29284/kesb4j60Keywords:
Neural Architecture Search, Meta- Learning, Domain Adaptation, Automated Deep Learning, Resource-Aware OptimizatiAbstract
Neural Architecture Search (NAS) has made its way as a powerful prototype for automating deep neural network design. However, conventional NAS approaches are specific to a single domain and computationally expensive, limiting their scalability across multiple domain environments such as computer vision, healthcare analytics, natural language processing, and edge intelligence. This paper introduces a Meta-Guided Domain- Aware NAS (MD-NAS) framework that forms an integration between meta-learning and domain-conditioned architecture generation. Nevertheless traditional NAS that searches independently for each task, the proposed method learns a shared meta- representation of architectural patterns that dynamically adapts to multi domain characteristics. A domain embedding module captures structural data properties and guides architecture sampling. Experimental results across multi-domain benchmarks demonstrate significant improvements in adaptation speed, generalization performance, and computational efficiency.
References
[1] Y. Hu, J. Liu, K. Wang, J. Zheng, W. Shi,
M. Zhang, Q. Dou, R. Liu, A. Zhou, and H. Li, “LM-Searcher: Cross-domain Neural Architecture
Search with LLMs via Unified Numerical Encoding,” EMNLP, 2025.
[2] X. Song, J. Xie, Y. Feng, A. Song, and
Y. Sun, “DAP: Domain Adaptive Performance Predictor for Efficient Neural Architecture Search,”
IEEE Trans. Computers, vol. 75, no. 1, pp. 152–163,
2026.
[3] V. Belagiannis, R. Asthana, J. Conrad,
Y. Dawoud, and M. Ortmanns, “Multi- conditioned Graph Diffusion for Neural Architecture
Search,” Transactions on Machine Learning Research (TMLR), 2024.
[4] S.-Y. Huang and W.-T. Chu, “Searching by Generating: Flexible and Efficient One- Shot NAS with
Architecture Generator,” in Proc. CVPR, 2021.
[5] R. Meng, C. Chen, W. Zhang, T. Qin, and T.-Y. Liu, “Slimmable Domain Adaptation,” in Proc.
CVPR, 2022.
[6] X. Deng, Y. Zhu, Y. Tian, and S. Newsam, “AutoAdapt: Automated Segmentation Network
Search for Unsupervised Domain Adaptation,” arXiv preprint arXiv:2106.13227, 2021.
[7] G. T. Pereira, “Neural Architecture Search with Interpretable Meta-Features,” Information
Sciences, 2023.
[8] S. Verma, “MLNAS: Meta-learning Based Neural Architecture Search (Two- Stage) for Unseen
Tasks,” J. Exp. Theor. Artificial Intelligence, 2024.
[9] A. Wang, K. Zhang, H. Wu, H. Chen, and M. Wang, “Meta-Learning-Integrated Neural
Architecture Search for Few-Shot Hyperspectral Image Classification,” Electronics, vol. 14, no. 15,
2025.
[10] Y. Liu et al., “Differentiable Neural Architecture Search for Domain Adaptation (Fault
Diagnosis Under Changing Conditions),” 2023.
[11] A. Heuillet, A. Nasser, H. Arioui, and
H. Tabia, “Efficient Automation of Neural Network Design: A Survey on Differentiable Neural
Architecture Search,” ACM Computing Surveys, 2024.
[12] S. M. Shariatzadeh et al., “A Survey on Multi-Objective Neural Architecture Search,” arXiv
preprint arXiv:2307.09099, 2023.
[13] X. Wang et al., “Advances in Neural Architecture Search,” National Science Review, 2024.
[14] R. Asthana et al., “Multi-conditioned Graph Diffusion for Neural Architecture Search,” arXiv
preprint arXiv:2403.06020, 2024.
[15] M. Pinos et al., “ApproxDARTS: Differentiable Neural Architecture Search for
Approximate/Hardware-Aware Deep Networks,” arXiv preprint arXiv:2404.08002, 2024.
[16] M. H. Rahman et al., “LEMONADE: An Automated Multi-Parameter Neural Architecture
Discovery Framework,” Scientific Reports, 2025.
[17] I. Legrene et al., “Enhancing Neural Architecture Search Using Transfer Learning and
Dynamic Search Space Adaptation,” Technologies (MDPI), 2025.
[18] O. Chakraborty, A. Sahoo, R. Panda, and A. Das, “ANYDA: Anytime Domain Adaptation with
Switchable Depth/Width/Resolution,” 2022.
[19] “Domain Adaptation One-Shot NAS (DAOS) with Network Similarity Directed Initialization
for Pathological Image Classification,” arXiv preprint, 2025.
[20] V. K. Chauhan et al., “A Brief Review of Hypernetworks in Deep Learning,” 2024.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
