DropNAS: Grouped Operation Dropout for Differentiable Architecture Search

Abstract

Neural architecture search (NAS) has shown encouraging results in automating the architecture design. Recently, DARTS relaxes the search process with a differentiable formulation that leverages weight-sharing and SGD where all candidate operations are trained simultaneously. Our empirical results show that such procedure results in the co-adaption problem and Matthew Effect: operations with fewer parameters would be trained maturely earlier. This causes two problems: firstly, the operations with more parameters may never have the chance to express the desired function since those with less have already done the job; secondly, the system will punish those underperforming operations by lowering their architecture parameter, and they will get smaller loss gradients, which causes the Matthew Effect. In this paper, we systematically study these problems and propose a novel grouped operation dropout algorithm named DropNAS to fix the problems with DARTS. Extensive experiments demonstrate that DropNAS solves the above issues and achieves promising performance. Specifically, DropNAS achieves 2.26% test error on CIFAR-10, 16.39% on CIFAR-100 and 23.4% on ImageNet (with the same training hyperparameters as DARTS for a fair comparison). It is also observed that DropNAS is robust across variants of the DARTS search space. Code is available at https://github.com/wiljohnhong/DropNAS.

References in corpus (2)

• Neural Architecture Search with Reinforcement Learning
• SNAS: Stochastic Neural Architecture Search

Cited by in corpus (11)

• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• Stronger NAS with Weaker Predictors
• DOTS: Decoupling Operation and Topology in Differentiable Architecture Search
• Auto-NBA: Efficient and Effective Search Over the Joint Space of Networks, Bitwidths, and Accelerators
• Single-level Optimization For Differential Architecture Search
• Interleaving Learning, with Application to Neural Architecture Search
• Learning by Self-Explanation, with Application to Neural Architecture Search
• Improving Differentiable Architecture Search with a Generative Model
• Single-DARTS: Towards Stable Architecture Search
• Learning from Mistakes based on Class Weighting with Application to Neural Architecture Search
• Small-Group Learning, with Application to Neural Architecture Search