论文标题
可再现体系结构改进的可区分神经体系结构转化
Differentiable Neural Architecture Transformation for Reproducible Architecture Improvement
论文作者
论文摘要
最近，引入了神经体系结构搜索（NAS）方法，并在许多基准测试中表现出令人印象深刻的性能。在NAS研究中，神经体系结构变压器（NAT）旨在改善给定的神经体系结构，在维持计算成本的同时具有更好的性能。但是，NAT对缺乏可重复性有局限性。在本文中，我们提出了可再现和高效的可区分神经体系结构变换。提出的方法在各种体系结构上显示出稳定的性能。在两个数据集（即CIFAR-10和Tiny Imagenet）上进行的广泛可重复性实验表明，该建议的方法绝对优于NAT，并且适用于其他模型和数据集。
Recently, Neural Architecture Search (NAS) methods are introduced and show impressive performance on many benchmarks. Among those NAS studies, Neural Architecture Transformer (NAT) aims to improve the given neural architecture to have better performance while maintaining computational costs. However, NAT has limitations about a lack of reproducibility. In this paper, we propose differentiable neural architecture transformation that is reproducible and efficient. The proposed method shows stable performance on various architectures. Extensive reproducibility experiments on two datasets, i.e., CIFAR-10 and Tiny Imagenet, present that the proposed method definitely outperforms NAT and be applicable to other models and datasets.
