A mutually exciting latent space Hawkes process model for continuous-time networks

Zhipeng Huang, Hadeel Soliman, Subhadeep Paul, Kevin S. Xu
Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, PMLR 180:863-873, 2022.

Abstract

Networks and temporal point processes serve as fundamental building blocks for modeling complex dynamic relational data in various domains. We propose the latent space Hawkes (LSH) model, a novel generative model for continuous-time networks of relational events, using a latent space representation for nodes. We model relational events between nodes using mutually exciting Hawkes processes with baseline intensities dependent upon the distances between the nodes in the latent space and sender and receiver specific effects. We demonstrate that our proposed LSH model can replicate many features observed in real temporal networks including reciprocity and transitivity, while also achieving superior prediction accuracy and providing more interpretable fits than existing models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v180-huang22b, title = {A mutually exciting latent space {Hawkes} process model for continuous-time networks}, author = {Huang, Zhipeng and Soliman, Hadeel and Paul, Subhadeep and Xu, Kevin S.}, booktitle = {Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence}, pages = {863--873}, year = {2022}, editor = {Cussens, James and Zhang, Kun}, volume = {180}, series = {Proceedings of Machine Learning Research}, month = {01--05 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v180/huang22b/huang22b.pdf}, url = {https://proceedings.mlr.press/v180/huang22b.html}, abstract = {Networks and temporal point processes serve as fundamental building blocks for modeling complex dynamic relational data in various domains. We propose the latent space Hawkes (LSH) model, a novel generative model for continuous-time networks of relational events, using a latent space representation for nodes. We model relational events between nodes using mutually exciting Hawkes processes with baseline intensities dependent upon the distances between the nodes in the latent space and sender and receiver specific effects. We demonstrate that our proposed LSH model can replicate many features observed in real temporal networks including reciprocity and transitivity, while also achieving superior prediction accuracy and providing more interpretable fits than existing models.} }
Endnote
%0 Conference Paper %T A mutually exciting latent space Hawkes process model for continuous-time networks %A Zhipeng Huang %A Hadeel Soliman %A Subhadeep Paul %A Kevin S. Xu %B Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2022 %E James Cussens %E Kun Zhang %F pmlr-v180-huang22b %I PMLR %P 863--873 %U https://proceedings.mlr.press/v180/huang22b.html %V 180 %X Networks and temporal point processes serve as fundamental building blocks for modeling complex dynamic relational data in various domains. We propose the latent space Hawkes (LSH) model, a novel generative model for continuous-time networks of relational events, using a latent space representation for nodes. We model relational events between nodes using mutually exciting Hawkes processes with baseline intensities dependent upon the distances between the nodes in the latent space and sender and receiver specific effects. We demonstrate that our proposed LSH model can replicate many features observed in real temporal networks including reciprocity and transitivity, while also achieving superior prediction accuracy and providing more interpretable fits than existing models.
APA
Huang, Z., Soliman, H., Paul, S. & Xu, K.S.. (2022). A mutually exciting latent space Hawkes process model for continuous-time networks. Proceedings of the Thirty-Eighth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 180:863-873 Available from https://proceedings.mlr.press/v180/huang22b.html.

Related Material