A Simple Latent Variable Model for Graph Learning and Inference

Manfred Jaeger, Antonio Longa, Steve Azzolin, Oliver Schulte, Andrea Passerini
Proceedings of the Second Learning on Graphs Conference, PMLR 231:26:1-26:18, 2024.

Abstract

We introduce a probabilistic latent variable model for graphs that generalizes both the established graphon and stochastic block models. This naive histogram AHK model is simple and versatile, and we demonstrate its use for disparate tasks including complex predictive inference usually not supported by other approaches, and graph generation. We analyze the tradeoffs entailed by the simplicity of the model, which imposes certain limitations on expressivity on the one hand, but on the other hand leads to robust generalization capabilities to graph sizes different from what was seen in the training data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v231-jaeger24a, title = {A Simple Latent Variable Model for Graph Learning and Inference}, author = {Jaeger, Manfred and Longa, Antonio and Azzolin, Steve and Schulte, Oliver and Passerini, Andrea}, booktitle = {Proceedings of the Second Learning on Graphs Conference}, pages = {26:1--26:18}, year = {2024}, editor = {Villar, Soledad and Chamberlain, Benjamin}, volume = {231}, series = {Proceedings of Machine Learning Research}, month = {27--30 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v231/jaeger24a/jaeger24a.pdf}, url = {https://proceedings.mlr.press/v231/jaeger24a.html}, abstract = {We introduce a probabilistic latent variable model for graphs that generalizes both the established graphon and stochastic block models. This naive histogram AHK model is simple and versatile, and we demonstrate its use for disparate tasks including complex predictive inference usually not supported by other approaches, and graph generation. We analyze the tradeoffs entailed by the simplicity of the model, which imposes certain limitations on expressivity on the one hand, but on the other hand leads to robust generalization capabilities to graph sizes different from what was seen in the training data.} }
Endnote
%0 Conference Paper %T A Simple Latent Variable Model for Graph Learning and Inference %A Manfred Jaeger %A Antonio Longa %A Steve Azzolin %A Oliver Schulte %A Andrea Passerini %B Proceedings of the Second Learning on Graphs Conference %C Proceedings of Machine Learning Research %D 2024 %E Soledad Villar %E Benjamin Chamberlain %F pmlr-v231-jaeger24a %I PMLR %P 26:1--26:18 %U https://proceedings.mlr.press/v231/jaeger24a.html %V 231 %X We introduce a probabilistic latent variable model for graphs that generalizes both the established graphon and stochastic block models. This naive histogram AHK model is simple and versatile, and we demonstrate its use for disparate tasks including complex predictive inference usually not supported by other approaches, and graph generation. We analyze the tradeoffs entailed by the simplicity of the model, which imposes certain limitations on expressivity on the one hand, but on the other hand leads to robust generalization capabilities to graph sizes different from what was seen in the training data.
APA
Jaeger, M., Longa, A., Azzolin, S., Schulte, O. & Passerini, A.. (2024). A Simple Latent Variable Model for Graph Learning and Inference. Proceedings of the Second Learning on Graphs Conference, in Proceedings of Machine Learning Research 231:26:1-26:18 Available from https://proceedings.mlr.press/v231/jaeger24a.html.

Related Material