PC-DARTS: Partial Channel Connections for Memory-Efficient Differentiable Architecture Search

12 Jul 2019Yuhui XuLingxi XieXiaopeng ZhangXin ChenGuo-Jun QiQi TianHongkai Xiong

Differentiable architecture search (DARTS) provided a fast solution in finding effective network architectures, but suffered from large memory and computing overheads in jointly training a super-network and searching for an optimal architecture. In this paper, we present a novel approach, namely, Partially-Connected DARTS, by sampling a small part of super-network to reduce the redundancy in exploring the network space, thereby performing a more efficient search without comprising the performance... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Neural Architecture Search CIFAR-10 PC-DARTS-CIFAR Top-1 Error Rate 2.51% # 2
Neural Architecture Search ImageNet PC-DARTS-ImagNet Top-1 Error Rate 24.2% # 1