A Dual Attention Network with Semantic Embedding for Few-Shot Learning
Shipeng Yan; Songyang Zhang; Xuming He
2019
Source PublicationTHIRTY-THIRD AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE / THIRTY-FIRST INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE / NINTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE
Pages9079-9086
AbstractDespite 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
Conference NameAAAI-2019
Conference PlaceHawaii USA
Conference Date2019-01
Indexed ByCPCI
Language英语
Funding ProjectNSFC[61703195]
WOS IDWOS:000486572503077
PublisherASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE
Original Document TypeProceedings Paper
Citation statistics
Cited Times [WOS]:0   [WOS Record]     [Related Records in WOS]
Document Type会议论文
Identifierhttp://kms.shanghaitech.edu.cn/handle/2MSLDSTB/29127
Collection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_硕士生
Corresponding AuthorXuming He
AffiliationShanghaiTech University
First Author AffilicationShanghaiTech University
Corresponding Author AffilicationShanghaiTech University
First Signature AffilicationShanghaiTech University
Recommended Citation
GB/T 7714
Shipeng Yan,Songyang Zhang,Xuming He. A Dual Attention Network with Semantic Embedding for Few-Shot Learning[C]:ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE,2019:9079-9086.
Files in This Item:
File Name/Size DocType Version Access License
stanet_aaai19_camera(1287KB)会议论文 限制开放CC BY-NC-SAView Application Full Text
Related Services
Usage statistics
Scholar Google
Similar articles in Scholar Google
[Shipeng Yan]'s Articles
[Songyang Zhang]'s Articles
[Xuming He]'s Articles
Baidu academic
Similar articles in Baidu academic
[Shipeng Yan]'s Articles
[Songyang Zhang]'s Articles
[Xuming He]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Shipeng Yan]'s Articles
[Songyang Zhang]'s Articles
[Xuming He]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: stanet_aaai19_camera (4).pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.