arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1902.09635cs.LGstat.ML
收藏

NAS-Bench-101:Towards Reproducible Neural Architecture搜索

NAS-Bench-101: Towards Reproducible Neural Architecture Search

Google Brain(谷歌大脑);University of Freiburg(弗莱堡大学)

Chris Ying, Aaron Klein, Esteban Real, Eric Christiansen, Kevin Murphy, Frank Hutter

摘要:

神经结构搜索(NAS)的最新进展需要巨大的计算资源,这使得很难重现实验,并对研究人员施加了进入大规模计算的障碍。我们的目标是通过引入NAS Bench-101来改善这些问题,NAS Bench-101是NAS研究的第一个公共架构数据集。为了构建NAS-Bench-101,我们仔细构建了一个紧凑但富有表现力的搜索空间,利用图同构来识别423 k个唯一的卷积架构。我们在CIFAR-10上对所有这些架构进行了多次训练和评估,并将结果编译成一个包含超过500万个训练模型的大型数据集。这使研究人员能够通过查询预先计算的数据集,在毫秒内评估各种模型的质量。我们展示了它的效用作为一个整体分析数据集,并通过基准测试的一系列架构优化算法。

英文摘要:

Recent advances in neural architecture search (NAS) demand tremendous computational resources, which makes it difficult to reproduce experiments and imposes a barrier-to-entry to researchers without access to large-scale computation. We aim to ameliorate these problems by introducing NAS-Bench-101, the first public architecture dataset for NAS research. To build NAS-Bench-101, we carefully constructed a compact, yet expressive, search space, exploiting graph isomorphisms to identify 423k unique convolutional architectures. We trained and evaluated all of these architectures multiple times on CIFAR-10 and compiled the results into a large dataset of over 5 million trained models. This allows researchers to evaluate the quality of a diverse range of models in milliseconds by querying the pre-computed dataset. We demonstrate its utility by analyzing the dataset as a whole and by benchmarking a range of architecture optimization algorithms.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑