GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders - IMAGINE
Communication Dans Un Congrès Année : 2018

GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders

Résumé

Deep learning on graphs has become a popular research topic with many applications. However, past work has concentrated on learning graph embedding tasks, which is in contrast with advances in generative models for images and text. Is it possible to transfer this progress to the domain of graphs? We propose to sidestep hurdles associated with linearization of such discrete structures by having a decoder output a probabilistic fully-connected graph of a predefined maximum size directly at once. Our method is formulated as a variational autoencoder. We evaluate on the challenging task of molecule generation.
Fichier principal
Vignette du fichier
icann_paper_long.pdf (384.19 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01990381 , version 1 (23-01-2019)

Identifiants

Citer

Martin Simonovsky, Nikos Komodakis. GraphVAE: Towards Generation of Small Graphs Using Variational Autoencoders. 27th International Conference on Artificial Neural Networks (ICANN), Oct 2018, Rhodes, Greece. ⟨10.1007/978-3-030-01418-6_41⟩. ⟨hal-01990381⟩
399 Consultations
987 Téléchargements

Altmetric

Partager

More