English

Exploring Complicated Search Spaces with Interleaving-Free Sampling

Machine Learning 2021-12-07 v1 Computer Vision and Pattern Recognition

Abstract

The existing neural architecture search algorithms are mostly working on search spaces with short-distance connections. We argue that such designs, though safe and stable, obstacles the search algorithms from exploring more complicated scenarios. In this paper, we build the search algorithm upon a complicated search space with long-distance connections, and show that existing weight-sharing search algorithms mostly fail due to the existence of \textbf{interleaved connections}. Based on the observation, we present a simple yet effective algorithm named \textbf{IF-NAS}, where we perform a periodic sampling strategy to construct different sub-networks during the search procedure, avoiding the interleaved connections to emerge in any of them. In the proposed search space, IF-NAS outperform both random sampling and previous weight-sharing search algorithms by a significant margin. IF-NAS also generalizes to the micro cell-based spaces which are much easier. Our research emphasizes the importance of macro structure and we look forward to further efforts along this direction.

Keywords

Cite

@article{arxiv.2112.02488,
  title  = {Exploring Complicated Search Spaces with Interleaving-Free Sampling},
  author = {Yunjie Tian and Lingxi Xie and Jiemin Fang and Jianbin Jiao and Qixiang Ye and Qi Tian},
  journal= {arXiv preprint arXiv:2112.02488},
  year   = {2021}
}

Comments

9 pages, 8 figures, 6 tables

R2 v1 2026-06-24T08:04:37.490Z