Algorithm 1 Classifier guided diffusion sampling, given a diffusion model $({\mu }_{\theta }\left(x_t\right),{\mathrm{\Sigma }}_{\theta }\left(x_t\right))$, classifier $p_{\varphi }\left(y\right|x_t)$, and gradient scale s

1:Input: class label y, gradient scale s 2:$x_T\leftarrow$ sample from $\mathcal{N}(0,\mathbf{I})$ 3:for all t from T to 1 do 4:   $\mu ,\mathrm{\Sigma }\leftarrow {\mu }_{\theta }\left(x_t\right),{\mathrm{\Sigma }}_{\theta }\left(x_t\right)$ 5:   $x_{t-1}\leftarrow$ sample from $\mathcal{N}(\mu +s\mathrm{\Sigma }{\nabla }_{x_t}log{p}_{\varphi }\left(y\right|x_t),\mathrm{\Sigma })$ 6:end for 7:return $x_0$