The COVID-19 pandemic has threatened global health. Many studies have applied deep convolutional neural networks (CNN) to recognize COVID-19 based on chest 3D computed tomography (CT). Recent works show that no model generalizes well across CT datasets from different countries, and manually designing models for specific datasets requires expertise; thus, neural architecture search (NAS) that aims to search models automatically has become an attractive solution. To reduce the search cost on large 3D CT datasets, most NAS-based works use the weight-sharing (WS) strategy to make all models share weights within a supernet; however, WS inevitably incurs search instability, leading to inaccurate model estimation. In this work, we propose an efficient Evolutionary Multi-objective ARchitecture Search (EMARS) framework. We propose a new objective, namely potential, which can help exploit promising models to indirectly reduce the number of models involved in weights training, thus alleviating search instability. We demonstrate that under objectives of accuracy and potential, EMARS can balance exploitation and exploration, i.e., reducing search time and finding better models. Our searched models are small and perform better than prior works on three public COVID-19 3D CT datasets.
翻译:COVID-19大流行已威胁到全球健康。许多研究应用了基于胸前3D计算断层成像仪(CT)的深演神经网络(CNN)来识别COVID-19。最近的工作显示,没有一种模型能够对不同国家的CT数据集进行全面概括的模型,而且人工设计具体数据集模型需要专门知识;因此,旨在自动搜索模型的神经结构搜索(NAS)已经成为一种有吸引力的解决方案。为了减少大型3DCT数据集的搜索成本,大多数NAS的工程都使用权重共享战略(WS)使所有模型在超级网络中共享重量;然而,WS不可避免地造成搜索不稳定,导致模型估计不准确。在这项工作中,我们提出了高效的进化多目标,即潜力,有助于利用有希望的模型间接减少参与重量培训的模型数量,从而缓解搜索不稳定性。我们证明,在精度和潜力目标下,EMARCT能够平衡开发和探索的所有模型,即D,从而减少搜索时间,并寻找更好的前三种模型。我们提出了一个新的目标,即潜力,即利用有希望的模型,从而间接地减少搜索三个公共模型。