中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Characterizing and Understanding Defense Methods for GNNs on GPUs

文献类型：期刊论文

作者Wu, Meng1,2; Yan, Mingyu2; Yang, Xiaocheng2; Li, Wenming2; Zhang, Zhimin2; Ye, Xiaochun2; Fan, Dongrui1,2
刊名IEEE COMPUTER ARCHITECTURE LETTERS
出版日期2023-07-01
卷号22期号:2页码:137-140
关键词Kernel Purification Estimation Graphics processing units Perturbation methods Electric breakdown Training Graph neural networks defense execution semantic execution pattern overhead
ISSN号1556-6056
DOI10.1109/LCA.2023.3304638
英文摘要Graph neural networks (GNNs) are widely deployed in many vital fields, but suffer from adversarial attacks, which seriously compromise the security in these fields. Plenty of defense methods have been proposed to mitigate the impact of these attacks, however, they have introduced extra time-consuming stages into the execution of GNNs. These extra stages need to be accelerated because the end-to-end acceleration is essential for GNNs to achieve fast development and deployment. To disclose the performance bottlenecks, execution patterns, execution semantics, and overheads of the defense methods for GNNs, we characterize and explore these extra stages on GPUs. Given the characterization and exploration, we provide several useful guidelines for both software and hardware optimizations to accelerate the defense methods for GNNs.
资助项目National Key Research and Development Program[2022YFB4501404] ; National Natural Science Foundation of China[62202451] ; CAS Project for Young Scientists in Basic Research[YSBR-029] ; CAS Project for Youth Innovation Promotion Association
WOS研究方向Computer Science
语种英语
WOS记录号WOS:001082420100001
出版者IEEE COMPUTER SOC
源URL[http://119.78.100.204/handle/2XEOYT63/21117]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Yan, Mingyu
作者单位1.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
2.Chinese Acad Sci, Inst Comp Technol, SKLP, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Wu, Meng,Yan, Mingyu,Yang, Xiaocheng,et al. Characterizing and Understanding Defense Methods for GNNs on GPUs[J]. IEEE COMPUTER ARCHITECTURE LETTERS,2023,22(2):137-140.
APA Wu, Meng.,Yan, Mingyu.,Yang, Xiaocheng.,Li, Wenming.,Zhang, Zhimin.,...&Fan, Dongrui.(2023).Characterizing and Understanding Defense Methods for GNNs on GPUs.IEEE COMPUTER ARCHITECTURE LETTERS,22(2),137-140.
MLA Wu, Meng,et al."Characterizing and Understanding Defense Methods for GNNs on GPUs".IEEE COMPUTER ARCHITECTURE LETTERS 22.2(2023):137-140.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。