Non Parametric Graph Learning for Bayesian Graph Neural Networks

Soumyasundar Pal, Saber Malekmohammadi, Florence Regol, Yingxue Zhang, Yishi Xu, Mark Coates
Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), PMLR 124:1318-1327, 2020.

Abstract

Graphs are ubiquitous in modelling relationalstructures. Recent endeavours in machine learningfor graph structured data have led to manyarchitectures and learning algorithms. However,the graph used by these algorithms is oftenconstructed based on inaccurate modellingassumptions and/or noisy data. As a result, itfails to represent the true relationships betweennodes. A Bayesian framework which targetsposterior inference of the graph by consideringit as a random quantity can be beneficial. Inthis paper, we propose a novel non-parametricgraph model for constructing the posterior distributionof graph adjacency matrices. The proposedmodel is flexible in the sense that it caneffectively take into account the output of graphbased learning algorithms that target specifictasks. In addition, model inference scales wellto large graphs. We demonstrate the advantagesof this model in three different problem settings:node classification, link prediction andrecommendation.

Cite this Paper

BibTeX
@InProceedings{pmlr-v124-pal20a, title = {Non Parametric Graph Learning for Bayesian Graph Neural Networks}, author = {Pal, Soumyasundar and Malekmohammadi, Saber and Regol, Florence and Zhang, Yingxue and Xu, Yishi and Coates, Mark}, booktitle = {Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI)}, pages = {1318--1327}, year = {2020}, editor = {Peters, Jonas and Sontag, David}, volume = {124}, series = {Proceedings of Machine Learning Research}, month = {03--06 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v124/pal20a/pal20a.pdf}, url = {https://proceedings.mlr.press/v124/pal20a.html}, abstract = {Graphs are ubiquitous in modelling relationalstructures. Recent endeavours in machine learningfor graph structured data have led to manyarchitectures and learning algorithms. However,the graph used by these algorithms is oftenconstructed based on inaccurate modellingassumptions and/or noisy data. As a result, itfails to represent the true relationships betweennodes. A Bayesian framework which targetsposterior inference of the graph by consideringit as a random quantity can be beneficial. Inthis paper, we propose a novel non-parametricgraph model for constructing the posterior distributionof graph adjacency matrices. The proposedmodel is flexible in the sense that it caneffectively take into account the output of graphbased learning algorithms that target specifictasks. In addition, model inference scales wellto large graphs. We demonstrate the advantagesof this model in three different problem settings:node classification, link prediction andrecommendation.} }
Endnote
%0 Conference Paper %T Non Parametric Graph Learning for Bayesian Graph Neural Networks %A Soumyasundar Pal %A Saber Malekmohammadi %A Florence Regol %A Yingxue Zhang %A Yishi Xu %A Mark Coates %B Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI) %C Proceedings of Machine Learning Research %D 2020 %E Jonas Peters %E David Sontag %F pmlr-v124-pal20a %I PMLR %P 1318--1327 %U https://proceedings.mlr.press/v124/pal20a.html %V 124 %X Graphs are ubiquitous in modelling relationalstructures. Recent endeavours in machine learningfor graph structured data have led to manyarchitectures and learning algorithms. However,the graph used by these algorithms is oftenconstructed based on inaccurate modellingassumptions and/or noisy data. As a result, itfails to represent the true relationships betweennodes. A Bayesian framework which targetsposterior inference of the graph by consideringit as a random quantity can be beneficial. Inthis paper, we propose a novel non-parametricgraph model for constructing the posterior distributionof graph adjacency matrices. The proposedmodel is flexible in the sense that it caneffectively take into account the output of graphbased learning algorithms that target specifictasks. In addition, model inference scales wellto large graphs. We demonstrate the advantagesof this model in three different problem settings:node classification, link prediction andrecommendation.
APA
Pal, S., Malekmohammadi, S., Regol, F., Zhang, Y., Xu, Y. & Coates, M.. (2020). Non Parametric Graph Learning for Bayesian Graph Neural Networks. Proceedings of the 36th Conference on Uncertainty in Artificial Intelligence (UAI), in Proceedings of Machine Learning Research 124:1318-1327 Available from https://proceedings.mlr.press/v124/pal20a.html.

Related Material