Zero-Shot Neural Architecture Search (NAS) approaches propose novel training-free metrics called zero-shot proxies to substantially reduce the search time compared to the traditional training-based NAS. Despite the success on image classification, the effectiveness of zero-shot proxies is rarely evaluated on complex vision tasks such as semantic segmentation and object detection. Moreover, existing zero-shot proxies are shown to be biased towards certain model characteristics which restricts their broad applicability. In this paper, we empirically study the bias of state-of-the-art (SOTA) zero-shot proxy ZiCo across multiple vision tasks and observe that ZiCo is biased towards thinner and deeper networks, leading to sub-optimal architectures. To solve the problem, we propose a novel bias correction on ZiCo, called ZiCo-BC. Our extensive experiments across various vision tasks (image classification, object detection and semantic segmentation) show that our approach can successfully search for architectures with higher accuracy and significantly lower latency on Samsung Galaxy S10 devices.

翻译：零样本神经架构搜索通过提出名为零样本代理的新型免训练度量指标，相较于传统基于训练的NAS方法大幅缩短了搜索时间。尽管在图像分类任务上取得成效，但零样本代理在语义分割、目标检测等复杂视觉任务中的有效性鲜有评估。此外，现有零样本代理被证实对特定模型特征存在偏差，这限制了其广泛适用性。本文针对当前最优零样本代理ZiCo在多种视觉任务中的偏差进行了实证研究，发现ZiCo倾向于选择更窄更深的网络，导致搜索到的架构次优。为解决该问题，我们提出一种针对ZiCo的新型偏差校正方法ZiCo-BC。通过在图像分类、目标检测和语义分割等多项视觉任务上的大量实验表明，我们的方法能够成功搜索出在三星Galaxy S10设备上具有更高精度且显著更低延迟的架构。