Event Uncertainty using Ensemble Neural Hawkes Process

Dubey, Manisha and Palakkadavath, Ragja and Srijith, P K (2023) Event Uncertainty using Ensemble Neural Hawkes Process. In: 6th ACM India Joint International Conference on Data Science and Management of Data, CODS-COMAD 2023, 4-7 January 2023, Mumbai.

[img] Text
3570991.3571002.pdf - Published Version

Download (490kB)

Abstract

Various real world applications in science and industry are often recorded over time as asynchronous event sequences. These event sequences comprise of the time of occurrence of events. Different applications including such event sequences are crime analysis, earthquake prediction, neural spiking train study, infectious disease prediction etc. A principled framework for modeling asynchronous event sequences is temporal point process. Recent works on neural temporal point process have combined the theoretical foundation of point process with universal approximation ability of neural networks. However, the predictions made by these models are uncertain due to incorrect model inference. Therefore, it is highly desirable to associate uncertainty with the predictions as well. In this paper, we propose a novel model, Ensemble Neural Hawkes Process, which is capable of predicting event occurrence time along with uncertainty, hence improving the generalization capability. We also propose evaluation metric which captures the uncertainty modelling capability for event prediction. The efficacy of proposed model is demonstrated using various simulated and real world datasets.

[error in script]
IITH Creators:
IITH CreatorsORCiD
Srijith, P Khttps://orcid.org/0000-0002-2820-0835
Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Event modeling; Hawkes process; Time-to-event prediction;Asynchronous event; Event model; Event prediction; Event sequence; Hawkes process; Point process; Real-world; Time to events; Time-to-event prediction; Uncertainty;Uncertainty analysis;Forecasting
Subjects: Computer science
Computer science > Computer programming, programs, data
Computer science > Big Data Analytics
Divisions: Department of Computer Science & Engineering
Depositing User: Mr Nigam Prasad Bisoyi
Date Deposited: 16 Aug 2023 11:14
Last Modified: 16 Aug 2023 11:14
URI: http://raiithold.iith.ac.in/id/eprint/11546
Publisher URL: https://doi.org/10.1145/3570991.3571002
Related URLs:

Actions (login required)

View Item View Item
Statistics for RAIITH ePrint 11546 Statistics for this ePrint Item