Knowledge Commons of Institute of Automation,CAS
Retaining Diverse Information in Contrastive Learning Through Multiple Projectors | |
He Zhu1,2![]() ![]() | |
发表期刊 | IEEE Signal Processing Letters
![]() |
2022-08-10 | |
卷号 | 29页码:1789-1793 |
摘要 | Contrastive Learning (CL) achieves great success in learning visual representations by comparing two augmented views of the same images. However, this very design removes transformation-dependent visual information from the pre-training, which leads to incomplete representations and is harmful for downstream tasks. It’s still an open question to retain such information in the CL pre-training process. In this paper, we propose a Multi-Projector Contrastive Learning (MPCL) to address this issue, which produces multi-view contrastive candidates to retain more comprehensive visual characteristics. In addition, we introduce a contrast regularization to construct multiple projectors as different as possible, thereby facilitating the diversity of preserved information. Finally, to promote a consistent learning process for multi-projector, we design a projector training balance strategy to adjust the learning preference of different projectors. MPCL can be applied to various CL frameworks to effectively protect visual characteristics. Experimental results show that the method performs well on subsequent tasks such as linear and semi-supervised image classification, object detection, and semantic segmentation. Importantly, the visual transformer trained by MPCL improves 2% absolute points of linear evaluation beyond the MoCo-v3 on the ImageNet-100 dataset. |
收录类别 | SCI |
语种 | 英语 |
七大方向——子方向分类 | 图像视频处理与分析 |
国重实验室规划方向分类 | 视觉信息处理 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/51646 |
专题 | 脑图谱与类脑智能实验室_脑网络组研究 |
作者单位 | 1.Brainnetome Center & National Laboratory of Pattern Recognition; Institute of Automation, Chinese Academy of Sciences; University of Chinese Academy of Sciences 2.School of Future Technology, University of Chinese Academy of Sciences(UCAS) |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | He Zhu,Shan Yu. Retaining Diverse Information in Contrastive Learning Through Multiple Projectors[J]. IEEE Signal Processing Letters,2022,29:1789-1793. |
APA | He Zhu,&Shan Yu.(2022).Retaining Diverse Information in Contrastive Learning Through Multiple Projectors.IEEE Signal Processing Letters,29,1789-1793. |
MLA | He Zhu,et al."Retaining Diverse Information in Contrastive Learning Through Multiple Projectors".IEEE Signal Processing Letters 29(2022):1789-1793. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Retaining_Diverse_In(1001KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
个性服务 |
推荐该条目 |
保存到收藏夹 |
查看访问统计 |
导出为Endnote文件 |
谷歌学术 |
谷歌学术中相似的文章 |
[He Zhu]的文章 |
[Shan Yu]的文章 |
百度学术 |
百度学术中相似的文章 |
[He Zhu]的文章 |
[Shan Yu]的文章 |
必应学术 |
必应学术中相似的文章 |
[He Zhu]的文章 |
[Shan Yu]的文章 |
相关权益政策 |
暂无数据 |
收藏/分享 |
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。
修改评论