DARTS: Differentiable Architecture Search
DARTS: Differentiable Architecture Search
复制标题
DOI：
--
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
Hanxiao Liu;K. Simonyan;Yiming Yang
中科院分区：
文献类型：
--
作者：
Hanxiao Liu;K. Simonyan;Yiming Yang
本文通过以可微分的方式制定任务来解决架构搜索的可扩展性挑战。与在离散且不可微的搜索空间上应用进化或强化学习的传统方法不同，我们的方法基于架构表示的连续松弛，允许使用梯度下降对架构进行有效搜索。在 CIFAR-10、ImageNet、Penn Treebank 和 WikiText-2 上进行的大量实验表明，我们的算法擅长发现用于图像分类的高性能卷积架构和用于语言建模的循环架构，同时比最先进的不可微分技术快几个数量级。我们的实现已公开发布，以促进对高效架构搜索算法的进一步研究。
This paper addresses the scalability challenge of architecture search by formulating the task in a differentiable manner. Unlike conventional approaches of applying evolution or reinforcement learning over a discrete and non-differentiable search space, our method is based on the continuous relaxation of the architecture representation, allowing efficient search of the architecture using gradient descent. Extensive experiments on CIFAR-10, ImageNet, Penn Treebank and WikiText-2 show that our algorithm excels in discovering high-performance convolutional architectures for image classification and recurrent architectures for language modeling, while being orders of magnitude faster than state-of-the-art non-differentiable techniques. Our implementation has been made publicly available to facilitate further research on efficient architecture search algorithms.