Fast Learning of Multidimensional Hawkes Processes via Frank-Wolfe

Renbo Zhao; Niccolò Dalmasso; Mohsen Ghassemi; Vamsi K Potluru; Tucker Balch; Manuela Veloso

doi:10.48550/arxiv.2212.06081

Back

Fast Learning of Multidimensional Hawkes Processes via Frank-Wolfe

Preprint

Open access

Fast Learning of Multidimensional Hawkes Processes via Frank-Wolfe

Renbo Zhao, Niccolò Dalmasso, Mohsen Ghassemi, Vamsi K Potluru, Tucker Balch and Manuela Veloso

ArXiv.org

Cornell University

12/12/2022

DOI: 10.48550/arxiv.2212.06081

Files and links (1)

url

https://doi.org/10.48550/arXiv.2212.06081View

Preprint (Author's original)This preprint has not been evaluated by subject experts through peer review. Preprints may undergo extensive changes and/or become peer-reviewed journal articles. Open Access

Abstract

Hawkes processes have recently risen to the forefront of tools when it comes to modeling and generating sequential events data. Multidimensional Hawkes processes model both the self and cross-excitation between different types of events and have been applied successfully in various domain such as finance, epidemiology and personalized recommendations, among others. In this work we present an adaptation of the Frank-Wolfe algorithm for learning multidimensional Hawkes processes. Experimental results show that our approach has better or on par accuracy in terms of parameter estimation than other first order methods, while enjoying a significantly faster runtime.

Machine Learning

Optimization and Control

Details

Title: Subtitle: Fast Learning of Multidimensional Hawkes Processes via Frank-Wolfe
Creators: Renbo Zhao
Niccolò Dalmasso
Mohsen Ghassemi
Vamsi K Potluru
Tucker Balch
Manuela Veloso
Resource Type: Preprint
Publication Details: ArXiv.org
DOI: 10.48550/arxiv.2212.06081
ISSN: 2331-8422
Publisher: Cornell University
Number of pages: 9
Comment: Presented at the NeurIPS 2022 Workshop on Synthetic Data for Empowering ML Research.
Language: English
Date posted: 12/12/2022
Academic Unit: Business Analytics
Record Identifier: 9984446692702771

Metrics

13 Record Views