Online Learning for Multivariate Hawkes Processes

Online Learning for Multivariate Hawkes Processes
复制标题

多元霍克斯过程的在线学习

DOI:
--
复制
发表时间:
2017
期刊:
Neural Information Processing Systems
影响因子:
--
通讯作者:
N. Kiyavash
N. Kiyavash
中科院分区:
--
文献类型:
--
作者:
Yingxiang Yang;Jalal Etesami;Niao He;N. Kiyavash

文献摘要

参考文献

被引文献

相似文献

提出了一种非参数在线学习算法,用于估计多变量Hawkes过程的触发函数。我们采用的方法通过再现核希尔伯特空间(RKHS)中的函数逼近触发函数$f_{i,j}(t)$,并使用Tikhonov正则化最大化对数似然的时间离散版本。从理论上讲,我们的算法实现了一个$\calO(\log T)$后悔界。数值结果表明,该算法具有与非参数批量学习算法相媲美的性能,其运行时间与参数在线学习算法相当。
We develop a nonparametric and online learning algorithm that estimates the triggering functions of a multivariate Hawkes process (MHP). The approach we take approximates the triggering function $f_{i,j}(t)$ by functions in a reproducing kernel Hilbert space (RKHS), and maximizes a time-discretized version of the log-likelihood, with Tikhonov regularization. Theoretically, our algorithm achieves an $\calO(\log T)$ regret bound. Numerical results show that our algorithm offers a competing performance to that of the nonparametric batch learning algorithm, with a run time comparable to the parametric online learning algorithm.
使用再生核进行泊松强度估计
DOI: 10.1214/17-ejs1339si
发表时间: 2017
影响因子: 1.1
作者:
Flaxman S
通讯作者: Flaxman S
DOI: 10.1016/j.ress.2004.07.017
发表时间: 2005-06-01
影响因子: 8.1
作者:
Gheorghe, AV;Birchmeier, J;Kröger, W
通讯作者: Kröger, W