Skip to main content
Cross Validated

Return to Question

added 10 characters in body
Source Link
User1865345
User1865345
  • 10.3k
  • 12
  • 23
  • 40

The forward diffusion process, which goes from x_t$x_t$ to x_{t+1}$x_{t+1} $ is Gaussian, which is very reasonable as we go the next state by adding random gaussian noise. However, I do not understand why the reverse of the process ( which goes from x_{t+1} $x_{t+1} $ to x_t $x_t $) is not also a Gaussian?.

The forward diffusion process, which goes from x_t to x_{t+1} is Gaussian, which is very reasonable as we go the next state by adding random gaussian noise. However, I do not understand why the reverse of the process ( which goes from x_{t+1} to x_t ) is not also a Gaussian?

The forward diffusion process, which goes from $x_t$ to $x_{t+1} $ is Gaussian, which is very reasonable as we go the next state by adding random gaussian noise. However, I do not understand why the reverse of the process ( which goes from $x_{t+1} $ to $x_t $) is not also a Gaussian.

Source Link
levitatmas
levitatmas
  • 325
  • 1
  • 6

Why reverse diffusion process is not a gaussian distribution?

The forward diffusion process, which goes from x_t to x_{t+1} is Gaussian, which is very reasonable as we go the next state by adding random gaussian noise. However, I do not understand why the reverse of the process ( which goes from x_{t+1} to x_t ) is not also a Gaussian?