目录
* Another Approach: q(z)->p(z)
<https://www.cnblogs.com/nickchen121/p/11073136.html#another-approach-qz-pz>
* Intuitively comprehend KL(p|q)
<https://www.cnblogs.com/nickchen121/p/11073136.html#intuitively-comprehend-klpq>
* Minimize KL Divergence
<https://www.cnblogs.com/nickchen121/p/11073136.html#minimize-kl-divergence>
* How to compute KL between q(z) and p(z)
<https://www.cnblogs.com/nickchen121/p/11073136.html#how-to-compute-kl-between-qz-and-pz>
* Distribution of hidden code
* Give more details after GAN
Another Approach: q(z)->p(z)
* Explicitly enforce
Intuitively comprehend KL(p|q)
\(KL \in{[0,+\infty]}\)
* KL = 0 --> \(p\approx{q}\)
Minimize KL Divergence
* Evidence Lower BOund
\[ KL(q_\theta(z|x_i)||p(z)) \]
How to compute KL between q(z) and p(z)
我们所希望的两点:
* 重建变量
* 隐藏变量逼近p的分布
热门工具 换一换