0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 1 year has passed since last update.

Deep Unsupervised Learning using Nonequilibrium Thermodynamicsの個人的理解用メモ

Last updated at Posted at 2023-04-17

論文情報

Deep Unsupervised Learning using Nonequilibrium Thermodynamics,Jascha Sohl-Dicksteinら,2015

論文の立ち位置

diffusionでよく引用されている比較的最初の論文

簡単な説明

  • マルコフ連鎖を使用してある分布を別の分布に徐々に変換

数式での説明(結果のみ)

Forward Trajectory

  • $q(x^{(0)})$:データ分布
  • $\pi(y)$:扱いやすい分布(正規分布とか)
  • $\beta$:拡散率
\pi(y)= \int dy'T_x(y|y';\beta)\pi(y')

y'からyへy徐々に変換していく。これはxでいうと、$x^{(t-1)}$から$x^{(t)}$へ徐々に変更していく

q(x^{(t)}|x^{(t-1)})=T_\pi(x^{(t)}|x^{(t-1)};\beta_t)

以下工事中

その他

内容に問題がある場合も責任は持てません。
コメントをいただければ修正します。

0
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?