论文解读(PAWS)《Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples》
论文信息
论文标题:Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples
论文作者:Mahmoud Assran, Mathilde Caron, Ishan Misra, Piotr Bojanowski, Armand Joulin, Nicolas Ballas
论文来源:NeurIPS 2021
论文地址:download
论文代码:download
视屏讲解:click
1 介绍
提出问题:充分利用无标注目标;
解决办法:对无标注数据使用一种可信的伪标签策略;
2 方法
2.1 整体框架
2.2 问题定义
We consider a large dataset of unlabeled images $\mathcal{D}= \left(\mathrm{x}_{i}\right)_{i \in[1, N]}$ and a small support dataset of annotated images $\mathcal{S}=\left(\mathbf{x}_{s i}, y_{i}\right)_{i \in[1, M]}$ , with $M \ll N $.Our goal is to learn image representations by leveraging both $\mathcal{D}$ and $\mathcal{S}$ during pretraining. After pre-training with $\mathcal{D}$ and $\mathcal{S}$ , we fine-tune the learned representations using only the labeled set $\mathcal{S}$ .
2.3 相似度分类器
公式:
$\pi_{d}\left(z_{i}, \mathbf{z}_{\mathcal{S}}\right)=\sum_{\left(z_{s_{j}}, y_{j}\right) \in \mathbf{z}_{\mathcal{S}}}\left(\frac{d\left(z_{i}, z_{s j}\right)}{\sum_{z_{s k} \in \mathbf{z}_{\mathcal{S}}} d\left(z_{i}, z_{s k}\right)}\right) y_{j}$
Note:Soft Nearest Neighbours strategy
其中:
$d(a, b) =\exp \left( \frac{a^{T} b}{\|a\|\|b\| \tau} \right)$
简化:
$p_{i}:=\pi_{d}\left(z_{i}, \mathbf{z}_{\mathcal{S}}\right)=\sigma_{\tau}\left(z_{i} \mathbf{z}_{\mathcal{S}}^{\top}\right) \mathbf{y}_{\mathcal{S}}$
Note:$p_{i} \in[0,1]^{K}$
锐化:
$\left[\rho\left(p_{i}\right)\right]_{k}:=\frac{\left[p_{i}\right]_{k}^{1 / T}}{\sum_{j=1}^{K}\left[p_{i}\right]_{j}^{1 / T}}, \quad k=1, \ldots, K$
Note:锐化目标会鼓励网络产生自信的预测,避免模型崩溃问题。
2.4 训练目标
总目标:
$\frac{1}{2 n} \sum_{i=1}^{n}\left(H\left(\rho\left(p_{i}^{+}\right), p_{i}\right)+H\left(\rho\left(p_{i}\right), p_{i}^{+}\right)\right)-H(\bar{p})$
ME-MAX 正则化项:$H(\bar{p})$
其中:
$\bar{p}:=\frac{1}{2 n} \sum_{i=1}^{n}\left(\rho\left(p_{i}\right)+\right. \left.\rho\left(p_{i}^{+}\right)\right) $ 表示所有未标记表示的锐化预测的平均值;
该正则化项在鼓励个体预测有信心的同时,鼓励平均预测接近均匀分布。ME-MAX 正则化项之前已在判别式无监督聚类社区中用于平衡学习的聚类大小[35]。
Note:交叉熵
# Example of target with class indices
loss = nn.CrossEntropyLoss()
input = torch.randn(3, 5, requires_grad=True)
target = torch.empty(3, dtype=torch.long).random_(5)
output = loss(input, target)
output.backward()
# Example of target with class probabilities
input = torch.randn(3, 5, requires_grad=True)
target = torch.randn(3, 5).softmax(dim=1)
output = loss(input, target)
output.backward()
3 总结
略
论文解读(PAWS)《Semi-Supervised Learning of Visual Features by Non-Parametrically Predicting View Assignments with Support Samples》的更多相关文章
- 【论文阅读】Deep Clustering for Unsupervised Learning of Visual Features
文章:Deep Clustering for Unsupervised Learning of Visual Features 作者:Mathilde Caron, Piotr Bojanowski, ...
- 论文解读《Momentum Contrast for Unsupervised Visual Representation Learning》俗称 MoCo
论文题目:<Momentum Contrast for Unsupervised Visual Representation Learning> 论文作者: Kaiming He.Haoq ...
- 论文解读《Deep Resdual Learning for Image Recognition》
总的来说这篇论文提出了ResNet架构,让训练非常深的神经网络(NN)成为了可能. 什么是残差? "残差在数理统计中是指实际观察值与估计值(拟合值)之间的差."如果回归模型正确的话 ...
- 【公式详解】【优秀论文解读】EDPLVO: Efficient Direct Point-Line Visual Odometry
前言 多的不说哈 2022最佳优秀论文 来自美团无人机团队 作者提出了一种使用点和线的高效的直接视觉里程计(visual odometry,VO)算法-- EDPLVO .他们证明了,2D 线上的 3 ...
- 论文解读(SimCLR)《A Simple Framework for Contrastive Learning of Visual Representations》
1 题目 <A Simple Framework for Contrastive Learning of Visual Representations> 作者: Ting Chen, Si ...
- 自监督学习(Self-Supervised Learning)多篇论文解读(下)
自监督学习(Self-Supervised Learning)多篇论文解读(下) 之前的研究思路主要是设计各种各样的pretext任务,比如patch相对位置预测.旋转预测.灰度图片上色.视频帧排序等 ...
- 自监督学习(Self-Supervised Learning)多篇论文解读(上)
自监督学习(Self-Supervised Learning)多篇论文解读(上) 前言 Supervised deep learning由于需要大量标注信息,同时之前大量的研究已经解决了许多问题.所以 ...
- 论文解读(BYOL)《Bootstrap Your Own Latent A New Approach to Self-Supervised Learning》
论文标题:Bootstrap Your Own Latent A New Approach to Self-Supervised Learning 论文方向:图像领域 论文来源:NIPS2020 论文 ...
- 论文笔记之:Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning
论文笔记之:Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning 2017-06-06 21: ...
- Deep Reinforcement Learning for Visual Object Tracking in Videos 论文笔记
Deep Reinforcement Learning for Visual Object Tracking in Videos 论文笔记 arXiv 摘要:本文提出了一种 DRL 算法进行单目标跟踪 ...
随机推荐
- Stream流相关方法
LIST<对象> 转换MAP 并根据某个字段分组 // 并根据某个字段分组,并做了归类 Map<String, List<User>> collect = user ...
- NET Core 部署IIS 碰到得问题解决(内托管模式超时、不允许得请求谓词、直接请求无响应、拒绝服务405)
web.config 配置说明 典型的web.confg 配置. 注意其中hostingModel模式和requestTimeout 进程内托管需要注意使用单独的应用程序池: 请求超时默认5分钟,出错 ...
- nRF52832出现“APP_UART_COMMUNICATION_ERROR”的错误的问题
在调试nRF52832的uart的过程中,发现调试信息会时不时打印"APP_UART_COMMUNICATION_ERROR"这个错误,看上去似乎毫无规律.查看SDK的相关说明,可 ...
- protobuf怎么处理java中的Object和Object[],protobuf的bytestring和object[]
如题,作者一开始也遇到了这个比较棘手的问题. 话不多说,直接说解决方案. 这里使用bytestring,如果是object[]的话则用repeated定义即可. 那么问题又来了,用这个类型怎么做到与j ...
- error check
#define SYSTEM_PRTCT_NOERR 0 #define SYSTEM_PRTCT_COVER (1 << 0) /* */#define SYSTEM_PRTCT_LPH ...
- Deepin20安装pip3和Pydev插件
1. 安装jdk(见前面) 2. 安装eclipse(见前面) 3. Deepin20里已经有python3.7.3和python2.7.16,我们可以直接使用3.7.3,也是默认的python3.还 ...
- RabbitMQ service does not start due to corrupted mnesia database in RSA Security Analytics
rabbitmq异常:mnesia database损坏
- 编译GO报错:cgo: C compiler
cgo: C compiler "gcc" not found: exec: "gcc": executable file not found in %PATH ...
- MySQL 导出数据结构 If you don't want to restore GTIDs, pass --set-gtid-purged=OFF.
应用场景MYSQL导出数据结构 Warning: A partial dump from a server that has GTIDs will by default include the GTI ...
- 2.27总结——JDBC学习
今天初步了解了Javaweb的JDBC,了解其基础语句,以及连接数据库的方式,但是自我感觉很抽象,实际上手仍有些困难,需要参考模板,增删改查目前进度在增和查,继续努力,争取本学期尽快跟上同学学习进度!