Nonparametric Hawkes Processes: Online Estimation and Generalization Bounds
This work addresses computational bottlenecks in event sequence modeling for applications like finance or social networks, representing an incremental improvement by adapting nonparametric methods to online settings.
The paper tackles the challenge of nonparametric online estimation for multivariate Hawkes processes by developing algorithms (NPOLE-MHP and NPOLE-MMHP) that achieve O(1/T) regret and stability, performing as well as optimal maximum likelihood estimation with runtime comparable to parametric online methods.
In this paper, we design a nonparametric online algorithm for estimating the triggering functions of multivariate Hawkes processes. Unlike parametric estimation, where evolutionary dynamics can be exploited for fast computation of the gradient, and unlike typical function learning, where representer theorem is readily applicable upon proper regularization of the objective function, nonparametric estimation faces the challenges of (i) inefficient evaluation of the gradient, (ii) lack of representer theorem, and (iii) computationally expensive projection necessary to guarantee positivity of the triggering functions. In this paper, we offer solutions to the above challenges, and design an online estimation algorithm named NPOLE-MHP that outputs estimations with a $\mathcal{O}(1/T)$ regret, and a $\mathcal{O}(1/T)$ stability. Furthermore, we design an algorithm, NPOLE-MMHP, for estimation of multivariate marked Hawkes processes. We test the performance of NPOLE-MHP on various synthetic and real datasets, and demonstrate, under different evaluation metrics, that NPOLE-MHP performs as good as the optimal maximum likelihood estimation (MLE), while having a run time as little as parametric online algorithms.