zoukankan      html  css  js  c++  java
  • 李宏毅 2018最新GAN课程 class 3 Theory behind GAN

     

     

     

     Too much limitation of Gaussian model. The images are too blurry. So any general model?

    But if PG(x;θ) is a neural network, it's impossible to calculate the likelihood. ????

     

     We don't know the formulation of PG and Pdata, so how to calculate the divergence???? ---->>> discriminator!!!

     

     

    V(G,D) = maximum the output if data come from Pdata, and maximum the output of data from PG. --- >>> this process is identical to a binary classifier

     

     

     

    不是minima 和 saddle point

    而是maxima

     

     

     

     

     

    even if L(G) is not differentiable (a Max operation), the derivative is computable.

     

     

    Tip:

    Train D as much as possible

    Train G only to a moderate level

     

      

     

    the results are actually similar......

      

     

     

    green point is true data, blue point is from genrator

     https://www.youtube.com/watch?v=ebMei6bYeWw

     

    Some one would argue that discriminator shouldn't initialize with the last discriminator, but the operation also sounds reasonable on another perspective.

    Some one in a paper shows that the performance increases if the samples come from current and past generators, although it may not sound reasonable.

  • 相关阅读:
    c 语言练习__去掉多余的空白字符_修正
    c 语言练习__求到N的阶乘的和。
    <<c 和指针 >> 部分笔记。
    AsciiDoc Markup Syntax Summary
    gdb 基本命令
    Ubuntu开启防火墙
    转载 jre精简
    windows 下指定jre运行java程序批处理指令
    虚拟机vmware的NAT网络
    liunx 端口权限
  • 原文地址:https://www.cnblogs.com/ecoflex/p/9040918.html
Copyright © 2011-2022 走看看