Neural Point Process for Learning Spatiotemporal Event Dynamics

Zihao Zhou, Xingyi Yang, Ryan Rossi, Handong Zhao, Rose Yu
Proceedings of The 4th Annual Learning for Dynamics and Control Conference, PMLR 168:777-789, 2022.

Abstract

Learning the dynamics of spatiotemporal events is a fundamental problem. Neural point processes enhance the expressivity of point process models with deep neural networks. However, most existing methods only consider temporal dynamics without spatial modeling. We propose Deep Spatiotemporal Point Process (DeepSTPP), a deep dynamics model that integrates spatiotemporal point processes. Our method is flexible, efficient, and can accurately forecast irregularly sampled events over space and time. The key construction of our approach is the nonparametric space-time intensity function, governed by a latent process. The intensity function enjoys closed-form integration for the density. The latent process captures the uncertainty of the event sequence. We use amortized variational inference to infer the latent process with deep networks. Using synthetic datasets, we validate our model can accurately learn the true intensity function. On real-world benchmark datasets, our model demonstrates superior performance over state-of-the-art baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v168-zhou22a, title = {Neural Point Process for Learning Spatiotemporal Event Dynamics}, author = {Zhou, Zihao and Yang, Xingyi and Rossi, Ryan and Zhao, Handong and Yu, Rose}, booktitle = {Proceedings of The 4th Annual Learning for Dynamics and Control Conference}, pages = {777--789}, year = {2022}, editor = {Firoozi, Roya and Mehr, Negar and Yel, Esen and Antonova, Rika and Bohg, Jeannette and Schwager, Mac and Kochenderfer, Mykel}, volume = {168}, series = {Proceedings of Machine Learning Research}, month = {23--24 Jun}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v168/zhou22a/zhou22a.pdf}, url = {https://proceedings.mlr.press/v168/zhou22a.html}, abstract = {Learning the dynamics of spatiotemporal events is a fundamental problem. Neural point processes enhance the expressivity of point process models with deep neural networks. However, most existing methods only consider temporal dynamics without spatial modeling. We propose Deep Spatiotemporal Point Process (DeepSTPP), a deep dynamics model that integrates spatiotemporal point processes. Our method is flexible, efficient, and can accurately forecast irregularly sampled events over space and time. The key construction of our approach is the nonparametric space-time intensity function, governed by a latent process. The intensity function enjoys closed-form integration for the density. The latent process captures the uncertainty of the event sequence. We use amortized variational inference to infer the latent process with deep networks. Using synthetic datasets, we validate our model can accurately learn the true intensity function. On real-world benchmark datasets, our model demonstrates superior performance over state-of-the-art baselines.} }
Endnote
%0 Conference Paper %T Neural Point Process for Learning Spatiotemporal Event Dynamics %A Zihao Zhou %A Xingyi Yang %A Ryan Rossi %A Handong Zhao %A Rose Yu %B Proceedings of The 4th Annual Learning for Dynamics and Control Conference %C Proceedings of Machine Learning Research %D 2022 %E Roya Firoozi %E Negar Mehr %E Esen Yel %E Rika Antonova %E Jeannette Bohg %E Mac Schwager %E Mykel Kochenderfer %F pmlr-v168-zhou22a %I PMLR %P 777--789 %U https://proceedings.mlr.press/v168/zhou22a.html %V 168 %X Learning the dynamics of spatiotemporal events is a fundamental problem. Neural point processes enhance the expressivity of point process models with deep neural networks. However, most existing methods only consider temporal dynamics without spatial modeling. We propose Deep Spatiotemporal Point Process (DeepSTPP), a deep dynamics model that integrates spatiotemporal point processes. Our method is flexible, efficient, and can accurately forecast irregularly sampled events over space and time. The key construction of our approach is the nonparametric space-time intensity function, governed by a latent process. The intensity function enjoys closed-form integration for the density. The latent process captures the uncertainty of the event sequence. We use amortized variational inference to infer the latent process with deep networks. Using synthetic datasets, we validate our model can accurately learn the true intensity function. On real-world benchmark datasets, our model demonstrates superior performance over state-of-the-art baselines.
APA
Zhou, Z., Yang, X., Rossi, R., Zhao, H. & Yu, R.. (2022). Neural Point Process for Learning Spatiotemporal Event Dynamics. Proceedings of The 4th Annual Learning for Dynamics and Control Conference, in Proceedings of Machine Learning Research 168:777-789 Available from https://proceedings.mlr.press/v168/zhou22a.html.

Related Material