Variational autoencoder(VAE) is a slightly more modern and interesting take on autoencoding. A Variational autoencoder(VAE) assumes that the source data has some sort of underlying probability ...

To address this issue, we propose a novel hyperspectral anomaly detection method based on a spatial-spectral joint mask variational autoencoder (VAE). By combining the probabilistic modeling ...

We propose a Crystal Diffusion Variational Autoencoder (CDVAE) that captures the physical inductive bias of material stability. By learning from the data distribution of stable materials, the decoder ...

Crucially, state-of-the-art methods often produce outputs that are not valid. We make the key observation that frequently, discrete data can be represented as a parse tree from a context-free grammar.