Figure 6.

Illustration of variational autoencoder (VAE) (Kingma and Welling, 2014). Encoder $\mathcal{Q}$ learns latent space of input data x. From the learned latent space, latent code z is sampled and input to decoder $\mathcal{P}$. Finally, decoder $\mathcal{P}$ reconstructs input data $\widehat{\text{x}}$.