This repository contains the PyTorch implementation of the paper: [HEP-NAS: Towards Efficient Few-shot Neural Architecture Search via Hierarchical Edge Partitioning](https://openreview.net/forum?id=E8SoR6yoYi&referrer=[Author Console](%2Fgroup%3Fid%3DAAAI.org%2F2025%2FConference%2FAuthors%23your-submissions)) in AAAI 2025. This code builds on GM-NAS published on ICLR 2022, we appreciate for their excellent work.
For experiments on NASBench-201 space, please refer to nasbench201/README.md
For experiments on DARTS space, please refer to sota/cnn/README.md
@inproceedings{li2025hep, title={HEP-NAS: Towards Efficient Few-shot Neural Architecture Search via Hierarchical Edge Partitioning}, author={Li, Jianfeng and Zhang, Jiawen and Wang, Feng and Ma, Lianbo}, booktitle={Proceedings of the AAAI Conference on Artificial Intelligence}, volume={39}, number={17}, pages={18314--18322}, year={2025} }