Permutation Equivariant Generative Adversarial Networks for Graphs

Yoann Boget; Magda Gregorova; Alexandros Kalousis

Permutation Equivariant Generative Adversarial Networks for Graphs

Machine Learning 2021-12-08 v1

Authors: Yoann Boget , Magda Gregorova , Alexandros Kalousis

Abstract

One of the most discussed issues in graph generative modeling is the ordering of the representation. One solution consists of using equivariant generative functions, which ensure the ordering invariance. After having discussed some properties of such functions, we propose 3G-GAN, a 3-stages model relying on GANs and equivariant functions. The model is still under development. However, we present some encouraging exploratory experiments and discuss the issues still to be addressed.

Keywords

generative adversarial network equivariant neural network graph generation

Cite

@article{arxiv.2112.03621,
  title  = {Permutation Equivariant Generative Adversarial Networks for Graphs},
  author = {Yoann Boget and Magda Gregorova and Alexandros Kalousis},
  journal= {arXiv preprint arXiv:2112.03621},
  year   = {2021}
}

Comments

ELLIS Machine Learning for Molecule Discovery Workshop. 5 pages + ref. + appendix

Related papers

View all related →