Algorithm 2: Sampling

1. $x_T\sim \mathcal{N}(0,\mathbf{I})$ 2. for $t=T,\dots ,1$ do 3. $z\sim \mathcal{N}(0,1)$ if $t>1$, else $\mathbf{z}=\mathbf{0}$ 4. $x_{t-1}={\displaystyle \frac{1}{\sqrt{{\alpha }_t}}}\left(x_t-{\displaystyle \frac{1-{\alpha }_t}{\sqrt{1-{\overline{\alpha }}_t}}}{\Phi }_{\theta }(x_t,t,f\left(b\right))\right)+{\sigma }_{t}z$ 5. end for 6. return $x_0$