Learning Registered Point Processes from Idiosyncratic Observations

Hongteng Xu, Lawrence Carin, Hongyuan Zha
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:5443-5452, 2018.

Abstract

A parametric point process model is developed, with modeling based on the assumption that sequential observations often share latent phenomena, while also possessing idiosyncratic effects. An alternating optimization method is proposed to learn a “registered” point process that accounts for shared structure, as well as “warping” functions that characterize idiosyncratic aspects of each observed sequence. Under reasonable constraints, in each iteration we update the sample-specific warping functions by solving a set of constrained nonlinear programming problems in parallel, and update the model by maximum likelihood estimation. The justifiability, complexity and robustness of the proposed method are investigated in detail, and the influence of sequence stitching on the learning results is examined empirically. Experiments on both synthetic and real-world data demonstrate that the method yields explainable point process models, achieving encouraging results compared to state-of-the-art methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-xu18b, title = {Learning Registered Point Processes from Idiosyncratic Observations}, author = {Xu, Hongteng and Carin, Lawrence and Zha, Hongyuan}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {5443--5452}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/xu18b/xu18b.pdf}, url = {https://proceedings.mlr.press/v80/xu18b.html}, abstract = {A parametric point process model is developed, with modeling based on the assumption that sequential observations often share latent phenomena, while also possessing idiosyncratic effects. An alternating optimization method is proposed to learn a “registered” point process that accounts for shared structure, as well as “warping” functions that characterize idiosyncratic aspects of each observed sequence. Under reasonable constraints, in each iteration we update the sample-specific warping functions by solving a set of constrained nonlinear programming problems in parallel, and update the model by maximum likelihood estimation. The justifiability, complexity and robustness of the proposed method are investigated in detail, and the influence of sequence stitching on the learning results is examined empirically. Experiments on both synthetic and real-world data demonstrate that the method yields explainable point process models, achieving encouraging results compared to state-of-the-art methods.} }
Endnote
%0 Conference Paper %T Learning Registered Point Processes from Idiosyncratic Observations %A Hongteng Xu %A Lawrence Carin %A Hongyuan Zha %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-xu18b %I PMLR %P 5443--5452 %U https://proceedings.mlr.press/v80/xu18b.html %V 80 %X A parametric point process model is developed, with modeling based on the assumption that sequential observations often share latent phenomena, while also possessing idiosyncratic effects. An alternating optimization method is proposed to learn a “registered” point process that accounts for shared structure, as well as “warping” functions that characterize idiosyncratic aspects of each observed sequence. Under reasonable constraints, in each iteration we update the sample-specific warping functions by solving a set of constrained nonlinear programming problems in parallel, and update the model by maximum likelihood estimation. The justifiability, complexity and robustness of the proposed method are investigated in detail, and the influence of sequence stitching on the learning results is examined empirically. Experiments on both synthetic and real-world data demonstrate that the method yields explainable point process models, achieving encouraging results compared to state-of-the-art methods.
APA
Xu, H., Carin, L. & Zha, H.. (2018). Learning Registered Point Processes from Idiosyncratic Observations. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:5443-5452 Available from https://proceedings.mlr.press/v80/xu18b.html.

Related Material