VGAE
A generative model based on graph structure in which an encoder maps node features and the adjacency matrix to parameters of distributions in a latent space, and a decoder reconstructs the adjacency matrix. It is used for unsupervised graph representation learning.
Timeline (last 90 days)
SCARCE adopts a two-stage process: in the first stage, a variational graph autoencoder (VGAE) performs warm-up via a graph reconstruction task; in the second stage, an adversarial confusion mechanism is introduced.
A variational graph autoencoder (VGAE) is a generative model based on graph structure, whose encoder maps node features and the adjacency matrix to distribution parameters in latent space (mean and log-variance), while the decoder reconstructs the adjacency matrix from the latent representation.
All Facts (2)
SCARCE adopts a two-stage process: in the first stage, a variational graph autoencoder (VGAE) performs warm-up via a graph reconstruction task; in the second stage, an adversarial confusion mechanism is introduced.
50%UnverifiedA variational graph autoencoder (VGAE) is a generative model based on graph structure, whose encoder maps node features and the adjacency matrix to distribution parameters in latent space (mean and log-variance), while the decoder reconstructs the adjacency matrix from the latent representation.
50%