A Dual Attention Network with Semantic Embedding for Few-Shot Learning
Shipeng Yan; Songyang Zhang; Xuming He
2019
会议录名称THIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE
页码9079-9086
发表状态已发表
摘要Despite recent success of deep neural networks, it remains challenging to efficiently learn new visual concepts from limited training data. To address this problem, a prevailing strategy is to build a meta-learner that learns prior knowledge on learning from a small set of annotated data. However, most of existing meta-learning approaches rely on a global representation of images and a meta-learner with complex model structures, which are sensitive to background clutter and difficult to interpret. We propose a novel meta-learning method for few-shot classification based on two simple attention mechanisms: one is a spatial attention to localize relevant object regions and the other is a task attention to select similar training data for label prediction. We implement our method via a dual-attention network and design a semantic-aware meta-learning loss to train the meta-learner network in an end-to-end manner. We validate our model on three few-shot image classification datasets with extensive ablative study, and our approach shows competitive performances over these datasets with fewer parameters. For facilitating the future research, code and data split are available: https://github.com/tonysy/STANet-PyTorch
会议录编者/会议主办者Association for the Advancement of Artificial Intelligence
关键词Classification (of information) Model structures Semantic Web Semantics Attention mechanisms Background clutter Classification datasets Competitive performance Global representation Limited training data Meta-learning approach Shot classification
会议名称33rd AAAI Conference on Artificial Intelligence, AAAI 2019, 31st Annual Conference on Innovative Applications of Artificial Intelligence, IAAI 2019 and the 9th AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2019
出版地2275 E BAYSHORE RD, STE 160, PALO ALTO, CA 94303 USA
会议地点Honolulu, HI, United states
会议日期January 27, 2019 - February 1, 2019
收录类别CPCI ; CPCI-S ; EI
语种英语
资助项目NSFC[61703195]
WOS记录号WOS:000486572503077
出版者ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE
EI入藏号20203509102371
EI主题词Deep neural networks
EI分类号461.4 Ergonomics and Human Factors Engineering ; 716.1 Information Theory and Signal Processing ; 723 Computer Software, Data Handling and Applications ; 903 Information Science ; 903.1 Information Sources and Analysis
原始文献类型Proceedings Paper
引用统计
文献类型会议论文
条目标识符https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/29127
专题信息科学与技术学院_硕士生
信息科学与技术学院_PI研究组_何旭明组
信息科学与技术学院_博士生
通讯作者Xuming He
作者单位ShanghaiTech University
第一作者单位上海科技大学
通讯作者单位上海科技大学
第一作者的第一单位上海科技大学
推荐引用方式
GB/T 7714
Shipeng Yan,Songyang Zhang,Xuming He. A Dual Attention Network with Semantic Embedding for Few-Shot Learning[C]//Association for the Advancement of Artificial Intelligence. 2275 E BAYSHORE RD, STE 160, PALO ALTO, CA 94303 USA:ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE,2019:9079-9086.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
个性服务
查看访问统计
谷歌学术
谷歌学术中相似的文章
[Shipeng Yan]的文章
[Songyang Zhang]的文章
[Xuming He]的文章
百度学术
百度学术中相似的文章
[Shipeng Yan]的文章
[Songyang Zhang]的文章
[Xuming He]的文章
必应学术
必应学术中相似的文章
[Shipeng Yan]的文章
[Songyang Zhang]的文章
[Xuming He]的文章
相关权益政策
暂无数据
收藏/分享
文件名: stanet_aaai19_camera (4).pdf
格式: Adobe PDF
此文件暂不支持浏览
所有评论 (0)
暂无评论
 

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。