【论文笔记】Decoupling Representation and Classifier for Long-Tailed Recognition
这一篇其实并不是提出什么新的东西,而且是做了点类似综述的技术调用实验。省流:T-normalization最好用
摘要
现状:Existing solutions usually involve class-balancing strategies, e.g. by loss re-weighting, data re-sampling, or transfer learning from head- to tail-classes, but most of them adhere to the scheme of jointly learning representations and classifiers.
做法:we decouple the learning procedure into representation learning and classification, and systematically explore how different balancing strategies affect them for long-tailed recognition.
结论:The findings are surprising: (1)data imbalance might not be an issue in learning high-quality representations; (2)with representations learned with the simplest instance-balanced sampling, it is also possible to achieve strong long-tailed recognition ability by adjusting only the classifier.
Classification For LONG-TAILED RECOGNITION
cRT(Classifier Re-training)
re-train the classifier with class-balanced sampling. That is, keeping the representations fixed, we ramdomly re-initialize and optimize the classifier weights W and b for a small number of epochs using class-balanced sampling.
NCM (Nearest Class Mean classifier)
- compute the mean feature representation for each class on the training set
- perform nearest neightbor search either using cosine similarity or the Euclidean distance computed on L2-normalized mean features.
- the cosine similarity alleviates the weight imbalance problem via its inherent normalization.
t-normalized(t-normalized classifier)
inspired: after joint training with instance-balanced sampling, the norms of the weights || wj || are correlated with the cardinality of the classes nj, while, after fine-tuning the classifiers using class-balanced sampling, the norms of the classifier weights tend to be more similar.
做法:adjusting the classifier weight norms directly through the following t-normalization procedure.
LWS(Learnable weight scaling)
Another way of interpreting t-normalization would be to think of it as a re-scaling of the magnitude for each classifier wi keeping the direction unchanged.(将T-norm 转化为一种对每个分类器权重大小的重新缩放)
Sampling Strategies
Instance-balanced sampling:the most common way of sampling data, where each training example has equal probability of being selected.
Class-balanced sampling:each class has an equal probability of being selected. One can see this as a two-stage sampling strategy, where first a class is selected uniformly from the set of classes, and then an instance from that class is subsequently uniformly sampled.
Square-root sampling
A number of variants of the previous sampling strategies have been explored.
Progressive-balanced sampling
This involves first using instance-balanced sampling for a number of epochs, and then class-balanced sampling for the last epochs.
Experiments
As illustrated in Fig.4, this yields a wider classfication boundary in feature space, allowing the classifier to have much higher accuracy on data-rich classes, but hurting data-scarce classes. t-normalized classifiers alleviate this issue to some extent by providing more balanced classifier weight magnitudes.
官方提供的代码链接: https://github.com/facebookresearch/classifier-balancing.
【论文笔记】Decoupling Representation and Classifier for Long-Tailed Recognition相关推荐
- Decoupling Representation and Classifier for Long-Tailed Recognition论文笔记
Decoupling Representation and Classifier for Long-Tailed Recognition论文笔记 Facebook出品,ICLR2020会议论文,官方代 ...
- 长尾分布系列论文解析(一)Decoupling Representation and Classifier for Long-Tailed Recognition
大纲 引言 分类问题中的长尾分布 采样策略 解耦训练 实验结果 总结 引言 看了挺多长尾分布的论文,从中获益匪浅,长尾分布的问题并不仅仅只局限于早期的分类问题之中,而是广泛存在于深度学习的多项任务之 ...
- 《Decoupling Representation and Classifier》笔记
Paper:<Decoupling Representation and Classifier for Long-tailed Recognition> Published at ICLR ...
- 长尾分布之DECOUPLING REPRESENTATION AND CLASSIFIER FOR LONG-TAILED RECOGNITION
原始文档:https://www.yuque.com/lart/papers/drggso ICLR 2020的文章. 针对长尾分布的分类问题提出了一种简单有效的基于re-sample范式的策略. 提 ...
- Decoupling Representation and Classifier for Long-Tailed Recognition
Decoupling Representation and Classifier for Long-Tailed Recognition [ICLR2020] 解耦论文GitHub 文章目录 Deco ...
- 论文笔记:Bag of Tricks for Long-Tailed Visual Recognition with Deep Convolutional Neural Networks
论文地址:http://www.lamda.nju.edu.cn/zhangys/papers/AAAI_tricks.pdf 代码地址:https://github.com/zhangyongshu ...
- 解耦表征与分类器:DECOUPLING REPRESENTATION AND CLASSIFIER FOR LONG-TAILED RECOGNITION(ICLR2020)
简介 论文出自ICLR2020,新加坡国立大学和 Facebook AI联合著作. 代码:https://github.com/facebookresearch/classifier-balancin ...
- [论文笔记]Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses
Decoupling Direction and Norm for Efficient Gradient-Based L2 Adversarial Attacks and Defenses(2019 ...
- 论文笔记 Bipartite Flat-Graph Network for Nested Named Entity Recognition | ACL2020
论文作者: 论文链接:https://arxiv.org/pdf/2005.00436.pdf 代码实现:https://github.com/cslydia/BiFlaG Abstract 作者为嵌 ...
最新文章
- 关于Linux中文件权限的探究
- 一份简单的在 Linux下编译及调试 C 代码的指南
- Big Data應用:以玩家意見之數據分析來探討何謂健康型線上遊戲(上)
- 初识莫队——小Z的袜子
- python网络爬虫实战 吕文翔_实战Python网络爬虫
- c语言中逗号是什么字符,C语言中的逗号操作符
- 索引书单(持续更新中)
- Bailian4002 谁是你的潜在朋友【暴力】
- python如何实现接口安全_利用pypy沙箱模式实现安全的开放式Python用户编程接口的方法与流程...
- PS 图像调整算法——自动色阶 (Auto Levels)
- nexus配置第三方库文件
- 干货分享! 20种数学建模方法!
- 【1】Python 视频文字识别提取 - Mp4转换成Mp3
- 磁力计简单水平较准算法
- 如何关闭mac的SIP
- Nature拟推出机器智能在线期刊Nature Machine Intelligence
- oppo手机硬件测试软件,OPPO怎么测试手机硬件,只需要做这一步,进入工程模式...
- 店铺流量突然下滑是怎么回事,我们应该从哪些角度来分析?
- [洛谷P2123]皇后游戏
- java1.8新特性之stream流式算法
热门文章
- 个人安全防范意识过低—血的教训!
- 测试 Wi-Fi、蓝牙性能
- 【突破二次元壁】手把手教你用AnimeGAN将风景图转换成宫崎骏动漫风
- Glider CodeForces - 1041D
- 程序设计与算法三~C++面向对象程序设计~北大郭炜MOOC学习笔记chapter1第一章(新标准C++程序设计)
- 爬虫:JS逆向之Webpack-乾坤大挪移
- 单源最短路径-迪杰斯拉特算法
- mysql 复制 错误 1007_mysql5.7.24Gtid下主从复制报错1007
- 新标日初级上册单词9----12
- JS实现马赛克图片效果