Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
ACM Computing Surveys2021Vol. 54(9), pp. 1–37
Citations Over TimeTop 10% of 2021 papers
Lingxi Xie, Xin Chen, Kaifeng Bi, Longhui Wei, Yuhui Xu, Lanfei Wang, Zhengsu Chen, An Xiao, Jianlong Chang, Xiaopeng Zhang, Qi Tian
Abstract
Neural architecture search (NAS) has attracted increasing attention. In recent years, individual search methods have been replaced by weight-sharing search methods for higher search efficiency, but the latter methods often suffer lower instability. This article provides a literature review on these methods and owes this issue to the optimization gap . From this perspective, we summarize existing approaches into several categories according to their efforts in bridging the gap, and we analyze both advantages and disadvantages of these methodologies. Finally, we share our opinions on the future directions of NAS and AutoML. Due to the expertise of the authors, this article mainly focuses on the application of NAS to computer vision problems.
Related Papers
- → ‘On the very spot’: In Defence of Battle(2012)2 cited
- → The Advertising of the Battle Creek Sanitarium Company.(1906)
- NEW RESEARCHES ON THE BATTLE OF CHENXIA & THE BATTLE OF GAIXIA(2006)
- On Battle of Tactical Psychology in an Anti-terrorism Specific Battle(2006)
- → Battle at Rača Đumruk (from 3rd September to 14th October, 1914)(2019)