Introducation

1. intruduce the conditional version of GANs, which can be constructed by simply feeding the data , y.

2. the CGANs can be used to learn a multi-modal model.

3.GANs in order to sidestep the difficulty of approximating many intractable probabilistic computations.(为了避免许多难以处理的概率计算的近似困难)

4. Adversarial nets have the advantages that Markov chains are never needed, only backpropagation is used to obtain gradients, no inference is required during learning,

and a wide variety of factors and interactions can easily be incorporated into the model.(多种因素和相互作用可以很容易地纳入模型)

5.马尔可夫链(Markov Chain),描述了一种状态序列,其每个状态值取决于前面有限个状态。一般来说,其核心是满足条件期望和平稳的分布,保证在计算过程中能够得到想要的概率分布。而我们考虑的生成模型恰好可能有以下两种情况:

  1. 输入一个随机分布的数据(例如一张黑白像素夹杂的噪音图),输出期望的数据(一张头像)

  2. 输入含有噪音的数据(在原有的图像上添加噪点或缺损),输出除去噪点或补完后的数据(完整的原始图像),这种情况下的模型也可以叫做任意去噪的自编码器。

无论是哪种情况,我们都希望从模型输出的数据y的概率分布尽可能逼近训练数据集的概率分布。但是让计算机生成一段音乐,或者一张有意义的图片,这个分布是非常复杂,很难求解的;即使通过马尔可夫链取样,得到了一个生成模型,我们最终也很难对这个模型的效果进行评估,因为生成的音乐到底好不好听,不同的人会得到不同的答案。

6. GANs can produce state of the art log-likehood estimate and realistic samples.

7. but

Related Work

1. the challeage of scaling models to accommodate an extremely large number of predicted output categories (调整模型以适应非常多的预测输出类别的挑战), to adress this problem by leveraging additional information such as using natural language corpora.and even a simple linear mapping from image feature-space to word-representation-space can improve. 

2. the challage of focusing on learning one-to-one mapping from input to output,but many interesting problems belong to a probabilistic one-to-many mapping.to adress this challege by using a conditional probabilistic generative model , for example, the input is taken to be the conditioning variable and the one-to-many mapping is instantiated(实例化)as a conditional predictive distribution.

Method

1. to specify that the G can capture the data distribution and the D can estimate the probability that a sample came from the training data rather than G.

2. the input is z,  G and D are both trained simultaneously. we adjust the parameters for G to minimize $log(1-D(G(z)))$ and adjust the parameters for D to minimize $log(D(X))$

Import Details -----Conditional Adversarial Nets

The training mechanism of CGANs.

1. GANs can be extended to a conditional model if both the G and D are conditioned on some extra information y.

2. y can be any kind of auxiliary information such as class label or data from other modalities.

3. feeding y into both discriminator and generator as additional input layer.

4. prior input noise and y are combined into joint hidden representation 对抗性训练框架允许在如何组成这种隐藏的表示方面具有相当大的灵活性。

5. In the discriminator and are presented as inputs and to a discriminative function (embodied x again by a MLP in this case).

The formula of a objective function :

The framework of CGANs:

Experiment 

 1. this paper trained a CGANs on MNIST images conditioned on their class labels, encoded as one-hot vectors.

For G:

both z and y are mapped to hidden layers with RELU, with layer sizes 200 and 1000 respectively, then combined hidden ReLu layer of dimensionality 1200.

For D: 

The discriminator maps to a maxout [6] layer with 240 units and 5 pieces, and to a maxout layer with 50 units and 5 pieces. Both of the hidden layers mapped to a joint maxout layer with 240 units and 4 pieces before being fed to the sigmoid layer

For Training:

and best estimate of log-likehood on the validation set was used as stopping point.(并以验证集的对数似然最优估计值作为停止点)。

 Summary

CGANs outperforms compared with original GANs, we can combine the class label or data from other modalities into the input of G and D, in order to achieve conditional probabilities distribution and controlling GANs.

 

CGANs的更多相关文章

  1. (转)Deep Learning Research Review Week 1: Generative Adversarial Nets

    Adit Deshpande CS Undergrad at UCLA ('19) Blog About Resume Deep Learning Research Review Week 1: Ge ...

  2. Unsupervised Image-to-Image Translation Networks --- Reading Writing

    Unsupervised Image-to-Image Translation Networks --- Reading Writing 2017.03.03 Motivations: most ex ...

  3. Face Aging with Conditional Generative Adversarial Network 论文笔记

    Face Aging with Conditional Generative Adversarial Network 论文笔记 2017.02.28  Motivation: 本文是要根据最新的条件产 ...

  4. #论文笔记# [pix2pixHD] High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

    Ting-Chun Wang, Ming-Yu Liu, Jun-Yan Zhu, Andrew Tao, Jan Kautz, and Bryan Catanzaro. "High-Res ...

  5. StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation - 1 - 多个域间的图像翻译论文学习

    Abstract 最近在两个领域上的图像翻译研究取得了显著的成果.但是在处理多于两个领域的问题上,现存的方法在尺度和鲁棒性上还是有所欠缺,因为需要为每个图像域对单独训练不同的模型.为了解决该问题,我们 ...

  6. CSAGAN:LinesToFacePhoto: Face Photo Generation from Lines with Conditional Self-Attention Generative Adversarial Network - 1 - 论文学习

    ABSTRACT 在本文中,我们探讨了从线条生成逼真的人脸图像的任务.先前的基于条件生成对抗网络(cGANs)的方法已经证明,当条件图像和输出图像共享对齐良好的结构时,它们能够生成视觉上可信的图像.然 ...

  7. Learning Face Age Progression: A Pyramid Architecture of GANs-1-实现人脸老化

    Learning Face Age Progression: A Pyramid Architecture of GANs Abstract 人脸年龄发展有着两个重要的需求,即老化准确性和身份持久性, ...

  8. AT指令集之Call

    1.//unsolicited result code,URC表示BP->AP+ESIPCPI:<call_id>,<dir>,<sip_msg_type>, ...

  9. mtk 的conferrence call建立流程

    (重点看main_log与) 抓mtk log: 1.*#*#82533284#*#*      进入抓log UI 2.*#*#825364#*#*      进入工程模式 3.进入"Lo ...

随机推荐

  1. 爬虫必看,每日JS逆向之爱奇艺密码加密,今天你练了吗?

    友情提示:优先在公众号更新,在博客园更新较慢,有兴趣的关注一下知识图谱与大数据公众号,本次目标是抠出爱奇艺passwd加密JS代码,如果你看到了这一篇,说明你对JS逆向感兴趣,如果是初学者,那不妨再看 ...

  2. st表、RMQ和LCA

    int lca(int x,int y) { if(de[x]<de[y]) swap(x,y); int d=de[x]-de[y]; for(int i=log2(d);i>=0;i- ...

  3. ch4inrulz: 1.0.1靶机渗透

    ch4inrulz: 1.0.1靶机渗透 扫描主机端口,还行啦四个开放的端口,8011和80端口都运行着web服务. 80端口下的robots.txt告诉我们什么都没有 在8011端口的apache服 ...

  4. Metasploit之漏洞利用( Metasploitable2)

    每个操作系统都会存在各种Bug,像Windows这样有版权的操作系统,微软公司会快速地开发针对这些Bug或漏洞的补丁,并为用户提供更新.全世界有大量的漏洞研究人员会夜以继日地发现.研究新的Bug,这些 ...

  5. Centos-配置网络或显示当前网络接口状态-ifconfig

    ifconfig 配置网络或显示当前网络接口状态,必须由root用户执行 相关选项 -a 显示所有网络接口信息,包括活动或非活动 -s 显示活动接口简要信息 -v 如果网卡接口出现错误则返回错误信息 ...

  6. kubernetes下jenkins实战maven项目编译构建

    关于kubernetes环境的jenkins集群 在kubernetes环境部署的jenkins集群,执行任务时会新建pod,任务完成后pod被销毁,架构如下所示: 在kubernetes搭建jenk ...

  7. Linux系统编程—信号集操作函数

    先来回顾一下未决信号集是怎么回事. 信号从产生到抵达目的地,叫作信号递达.而信号从产生到递达的中间状态,叫作信号的未决状态.产生未决状态的原因有可能是信号受到阻塞了,也就是信号屏蔽字(或称阻塞信号集, ...

  8. (转)DBC文件格式解析

    Dbc是描述CAN通信报文和信号信息的文件,用Vector Candb++打开. 用记事本打开后,可以看到固定格式,下面的博客做了详细的解析: https://blog.csdn.net/weixin ...

  9. 【题解】Tree

    题目戳我 \(\text{Solution:}\) 考虑点分治.对于这个两点之间,它意味着这点对必须是不一样的. 考虑用双指针统计答案.显然,对于两个数\(a,b\),要让\(a+b=k,a\)越大则 ...

  10. 【题解】[USACO08MAR]Land Acquisition G

    Link 题目大意:给定\(n\)个二元组,每次可以选择一组,花费是组内最大的长乘以最大的宽.问消掉所有二元组的最小代价. \(\text{Solution:}\) \(dp\)写的不够啊-- 先挖掘 ...