Searching for A Robust Neural Architecture in Four GPU Hours

CVPR 2019 Xuanyi DongYi Yang

Conventional neural architecture search (NAS) approaches are based on reinforcement learning or evolutionary strategy, which take more than 3000 GPU hours to find a good model on CIFAR-10. We propose an efficient NAS approach learning to search by gradient descent... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Neural Architecture Search CIFAR-10 GDAS (FRC) Top-1 Error Rate 2.5 # 1
Neural Architecture Search CIFAR-10 GDAS (FRC) Search Time (GPU days) 0.17 # 1
Neural Architecture Search CIFAR-10 GDAS Top-1 Error Rate 3.4 # 4
Neural Architecture Search CIFAR-10 GDAS Search Time (GPU days) 0.21 # 2