当前位置:网站首页>Icml2022 | interventional contrastive learning based on meta semantic regularization
Icml2022 | interventional contrastive learning based on meta semantic regularization
2022-07-01 21:51:00 【Zhiyuan community】
Thesis link :https://arxiv.org/abs/2206.14702
Based on comparative learning (CL) The self supervised learning model of learning visual representations in pairs . Although the current popular CL The model has made great progress , But in this article , We found a phenomenon that has been neglected : When using full image training CL Model time , The performance tested in the full image is better than that tested in the foreground area ; When using foreground area training CL Model time , The performance tested in the complete image is worse than that tested in the foreground area . This observation shows that , The background in the image may interfere with the semantic information of model learning , Its influence has not been completely eliminated . To solve this problem , We build a structural causal model (SCM), Modeling the background as a confounding agent . We propose a regularization method based on backdoor adjustment , That is, interventional contrastive learning based on meta semantic regularization (ICLMSR), Conduct causal intervention on the proposed supply chain management .ICL-MSR Can be integrated into any existing CL In the method , To reduce the background interference of representation learning . It is proved theoretically ICL-MSR It has a smaller error bound . Empirically , Our experiments on several benchmark datasets show that ,ICL-MSR Can improve the most advanced CL Method performance .
边栏推荐
- 面试题:MySQL的union all和union有什么区别、MySQL有哪几种join方式(阿里面试题)[通俗易懂]
- js数组拼接的四种方法[通俗易懂]
- Significance and measures of security encryption of industrial control equipment
- PCB线路板塞孔工艺的那些事儿~
- 物联网rfid等
- 杰理之、产线装配环节【篇】
- Wechat applet, continuously playing multiple videos. Synthesize the appearance of a video and customize the video progress bar
- TOPS,处理器运算能力单位、每秒钟可进行一万亿次
- AirServer2022最新版功能介绍及下载
- leetcode刷题:栈与队列03(有效的括号)
猜你喜欢
随机推荐
Wechat applet, continuously playing multiple videos. Synthesize the appearance of a video and customize the video progress bar
基于YOLOv5的口罩佩戴检测方法
选择在同花顺上炒股开户可以吗?安全吗?
Can I choose to open an account for stock trading on flush? Is it safe?
都能看懂的LIS(最长上升子序列)问题[通俗易懂]
pytest合集(2)— pytest运行方式
Flume面试题
locust 系列入门
以飞地园区为样本,看雨花与韶山如何奏响长株潭一体化发展高歌
Slf4j打印异常的堆栈信息
基于LSTM模型实现新闻分类
杰理之蓝牙耳机品控和生产技巧【篇】
leetcode刷题:栈与队列01(用栈实现队列)
功利点没啥!
东哥套现,大佬隐退?
Halcon知识:三维重构的一个尝试
MySQL系列之事务日志Redo log学习笔记
深度学习 常见的损失函数
2022熔化焊接与热切割上岗证题目模拟考试平台操作
股票手机开户哪个app好,安全性较高的