English

Unifying and Boosting Gradient-Based Training-Free Neural Architecture Search

Machine Learning 2022-10-13 v2 Artificial Intelligence

Abstract

Neural architecture search (NAS) has gained immense popularity owing to its ability to automate neural architecture design. A number of training-free metrics are recently proposed to realize NAS without training, hence making NAS more scalable. Despite their competitive empirical performances, a unified theoretical understanding of these training-free metrics is lacking. As a consequence, (a) the relationships among these metrics are unclear, (b) there is no theoretical interpretation for their empirical performances, and (c) there may exist untapped potential in existing training-free NAS, which probably can be unveiled through a unified theoretical understanding. To this end, this paper presents a unified theoretical analysis of gradient-based training-free NAS, which allows us to (a) theoretically study their relationships, (b) theoretically guarantee their generalization performances, and (c) exploit our unified theoretical understanding to develop a novel framework named hybrid NAS (HNAS) which consistently boosts training-free NAS in a principled way. Remarkably, HNAS can enjoy the advantages of both training-free (i.e., the superior search efficiency) and training-based (i.e., the remarkable search effectiveness) NAS, which we have demonstrated through extensive experiments.

Keywords

Cite

@article{arxiv.2201.09785,
  title  = {Unifying and Boosting Gradient-Based Training-Free Neural Architecture Search},
  author = {Yao Shu and Zhongxiang Dai and Zhaoxuan Wu and Bryan Kian Hsiang Low},
  journal= {arXiv preprint arXiv:2201.09785},
  year   = {2022}
}

Comments

Published as a conference paper at NeurIPS 2022

R2 v1 2026-06-24T09:00:31.796Z