AutoML: A survey of the state-of-the-art
AutoML: A survey of the state-of-the-art
复制标题
DOI：
10.1016/j.knosys.2020.106622
复制
发表时间：
2021-01-05
影响因子：
8.8
通讯作者：
Chu, Xiaowen
中科院分区：
文献类型：
--
作者：
He, Xin;Zhao, Kaiyong;Chu, Xiaowen
深度学习技术在图像识别、目标检测、语言建模等方面取得了令人瞩目的成就。然而，针对特定任务构建高质量的数字图书馆系统高度依赖于人的专业知识，阻碍了其广泛应用。与此同时，自动机器学习(AutoML)是一种很有前途的解决方案，可以在没有人工帮助的情况下构建数字图书馆系统，并正在得到广泛的研究。本文对AutoML中的最新技术(SOTA)进行了全面和最新的回顾。针对目前AutoML的一个热点子课题--数据准备、特征工程、超参数优化和神经体系结构搜索(NAS)，我们介绍了AutoML方法，并重点介绍了NAS。我们总结了典型的NAS算法在CIFAR-10和ImageNet数据集上的性能，并进一步讨论了NAS方法的以下主题：一阶段/两阶段NAS、单次NAS、联合超参数和体系结构优化以及资源感知NAS。最后，我们讨论了与现有AutoML方法相关的一些有待进一步研究的问题。(C)2020爱思唯尔B.V.保留所有权利。
Deep learning (DL) techniques have obtained remarkable achievements on various tasks, such as image recognition, object detection, and language modeling. However, building a high-quality DL system for a specific task highly relies on human expertise, hindering its wide application. Meanwhile, automated machine learning (AutoML) is a promising solution for building a DL system without human assistance and is being extensively studied. This paper presents a comprehensive and up-to-date review of the state-of-the-art (SOTA) in AutoML. According to the DL pipeline, we introduce AutoML methods - covering data preparation, feature engineering, hyperparameter optimization, and neural architecture search (NAS) - with a particular focus on NAS, as it is currently a hot sub-topic of AutoML. We summarize the representative NAS algorithms' performance on the CIFAR-10 and ImageNet datasets and further discuss the following subjects of NAS methods: one/two-stage NAS, one-shot NAS, joint hyperparameter and architecture optimization, and resource-aware NAS. Finally, we discuss some open problems related to the existing AutoML methods for future research. (C) 2020 Elsevier B.V. All rights reserved.