I have been reading about how we can model a Variational AutoEncoder (VAE) into a Graph Variational AutoEncoder (GVAE) where the decoder reconstructs the adjacency matrix.

I presume that this approach makes the GVAE completely transductive because an unseen node will not be present in the adjacency matrix during training and we cannot dynamically change the dimensions of the matrix either. If we had to make it inductive then what approach can be taken? Is there already any work done in this regards? I am unable to find any appropriate literature so will really appreciate some help here.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.