引用本文: | 丁丁,刘文哲,盛常冲,等.神经网络架构搜索研究进展与展望[J].国防科技大学学报,2023,45(6):100-131.[点击复制] |
DING Ding,LIU Wenzhe,SHENG Changchong,et al.State of the art and prospects of neural architecture search[J].Journal of National University of Defense Technology,2023,45(6):100-131[点击复制] |
|
|
|
本文已被:浏览 2903次 下载 2706次 |
神经网络架构搜索研究进展与展望 |
丁丁1,刘文哲2,盛常冲3,隋金坪4,刘丽2 |
(1. 国防科技大学 教研保障中心, 湖南 长沙 410073;2. 国防科技大学 系统工程学院, 湖南 长沙 410073;3. 海军工程大学 电磁能技术全国重点实验室, 湖北 武汉 430033;4. 海军大连舰艇学院 作战软件与仿真研究所, 辽宁 大连 116016)
|
摘要: |
神经网络架构搜索旨在针对不同任务,自动化地搜索得到性能最优的神经网络结构,是深度学习、计算机视觉技术结合当前现实需求应运而生的一大重要科学问题。对近年来神经网络架构搜索研究进行梳理、归类和评述;阐述神经网络架构搜索的定义和意义,全方位剖析当前研究所面临的难点与挑战;以此为基础,对主流的搜索策略进行阐述和归纳;探讨研究潜在的问题及未来颇具潜力的研究方向,以期推动该领域的进一步发展。 |
关键词: 深度学习 神经网络架构搜索 自动机器学习 强化学习 搜索空间设计 搜索策略 进化算法 |
DOI:10.11887/j.cn.202306014 |
投稿日期:2022-02-19 |
基金项目:国家自然科学基金资助项目(61872379) |
|
State of the art and prospects of neural architecture search |
DING Ding1, LIU Wenzhe2, SHENG Changchong3, SUI Jinping4, LIU Li2 |
(1. Center for Teaching and Research Support, National University of Defense Technology, Changsha 410073, China;2. College of Systems Engineering, National University of Defense Technology, Changsha 410073, China;3. National Key Laboratory of Electromagnetic Energy, Naval University of Engineering, Wuhan 430033, China;4. Operational Software and Simulation Institute, Dalian Navy Academy, Dalian 116016)
|
Abstract: |
Neural architecture search is a task that aims to automatically search for the optimal neural network structure for different tasks, which is of great importance and inevitability in the joint development of deep learning and computer vision to the current stage. A comprehensive review of the research on neural network search was provided. In specific, the definition and significance of neural architecture search were introduced, and the difficulties and challenges faced in relevant research were deeply analyzed. Based on this, the mainstream search strategies was elaborate and summarize; Finally, the potential problems and possible future research directions were summarized and discussed to promote further development in this field. |
Keywords: deep learning neural architecture search automation of machine learning reinforcement learning search space design search strategy evolutionary algorithms |
|
|
|
|
|