Graphite: Iterative Generative Modeling of Graphs

Aditya Grover, Aaron Zweig, Stefano Ermon
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:2434-2444, 2019.

Abstract

Graphs are a fundamental abstraction for modeling relational data. However, graphs are discrete and combinatorial in nature, and learning representations suitable for machine learning tasks poses statistical and computational challenges. In this work, we propose Graphite, an algorithmic framework for unsupervised learning of representations over nodes in large graphs using deep latent variable generative models. Our model parameterizes variational autoencoders (VAE) with graph neural networks, and uses a novel iterative graph refinement strategy inspired by low-rank approximations for decoding. On a wide variety of synthetic and benchmark datasets, Graphite outperforms competing approaches for the tasks of density estimation, link prediction, and node classification. Finally, we derive a theoretical connection between message passing in graph neural networks and mean-field variational inference.

Cite this Paper


BibTeX
@InProceedings{pmlr-v97-grover19a, title = {Graphite: Iterative Generative Modeling of Graphs}, author = {Grover, Aditya and Zweig, Aaron and Ermon, Stefano}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {2434--2444}, year = {2019}, editor = {Kamalika Chaudhuri and Ruslan Salakhutdinov}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/grover19a/grover19a.pdf}, url = { http://proceedings.mlr.press/v97/grover19a.html }, abstract = {Graphs are a fundamental abstraction for modeling relational data. However, graphs are discrete and combinatorial in nature, and learning representations suitable for machine learning tasks poses statistical and computational challenges. In this work, we propose Graphite, an algorithmic framework for unsupervised learning of representations over nodes in large graphs using deep latent variable generative models. Our model parameterizes variational autoencoders (VAE) with graph neural networks, and uses a novel iterative graph refinement strategy inspired by low-rank approximations for decoding. On a wide variety of synthetic and benchmark datasets, Graphite outperforms competing approaches for the tasks of density estimation, link prediction, and node classification. Finally, we derive a theoretical connection between message passing in graph neural networks and mean-field variational inference.} }
Endnote
%0 Conference Paper %T Graphite: Iterative Generative Modeling of Graphs %A Aditya Grover %A Aaron Zweig %A Stefano Ermon %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-grover19a %I PMLR %P 2434--2444 %U http://proceedings.mlr.press/v97/grover19a.html %V 97 %X Graphs are a fundamental abstraction for modeling relational data. However, graphs are discrete and combinatorial in nature, and learning representations suitable for machine learning tasks poses statistical and computational challenges. In this work, we propose Graphite, an algorithmic framework for unsupervised learning of representations over nodes in large graphs using deep latent variable generative models. Our model parameterizes variational autoencoders (VAE) with graph neural networks, and uses a novel iterative graph refinement strategy inspired by low-rank approximations for decoding. On a wide variety of synthetic and benchmark datasets, Graphite outperforms competing approaches for the tasks of density estimation, link prediction, and node classification. Finally, we derive a theoretical connection between message passing in graph neural networks and mean-field variational inference.
APA
Grover, A., Zweig, A. & Ermon, S.. (2019). Graphite: Iterative Generative Modeling of Graphs. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:2434-2444 Available from http://proceedings.mlr.press/v97/grover19a.html .

Related Material