Fully Neural Network based Model for General Temporal Point Processes

Abstract

A temporal point process is a mathematical model for a time series of discrete events, which covers various applications. Recently, recurrent neural network (RNN) based models have been developed for point processes and have been found effective. RNN based models usually assume a specific functional form for the time course of the intensity function of a point process (e.g., exponentially decreasing or increasing with the time since the most recent event). However, such an assumption can restrict the expressive power of the model. We herein propose a novel RNN based model in which the time course of the intensity function is represented in a general manner. In our approach, we first model the integral of the intensity function using a feedforward neural network and then obtain the intensity function as its derivative. This approach enables us to both obtain a flexible model of the intensity function and exactly evaluate the log-likelihood function, which contains the integral of the intensity function, without any numerical approximations. Our model achieves competitive or superior performances compared to the previous state-of-the-art methods for both synthetic and real datasets.

References in corpus (4)

• Adam: A Method for Stochastic Optimization
• Deep Reinforcement Learning of Marked Temporal Point Processes
• Learning Temporal Point Processes via Reinforcement Learning
• Neural Likelihoods via Cumulative Distribution Functions

Cited by in corpus (19)

• Spatio-Temporal Graph Neural Point Process for Traffic Congestion Event Prediction
• Learning Neural Event Functions for Ordinary Differential Equations
• Intensity-Free Learning of Temporal Point Processes
• Self-Attentive Hawkes Processes
• TIGGER: Scalable Generative Modelling for Temporal Interaction Graphs
• Identifying Cost-effective Debunkers for Multi-stage Fake News Mitigation Campaigns
• Neural Temporal Point Processes For Modelling Electronic Health Records
• Fast and Flexible Temporal Point Processes with Triangular Maps
• Spatio-Temporal Point Processes with Attention for Traffic Congestion Event Modeling
• An Empirical Study: Extensive Deep Temporal Point Process
• Deep Fourier Kernel for Self-Attentive Point Processes
• Semi-supervised Learning for Marked Temporal Point Processes
• Neural Spectral Marked Point Processes
• Identifying Coordinated Accounts on Social Media through Hidden Influence and Group Behaviours
• UNIPoint: Universally Approximating Point Processes Intensities
• A Metric Space for Point Process Excitations
• Fine-Grained $\epsilon$-Margin Closed-Form Stabilization of Parametric Hawkes Processes
• Mitigating Performance Saturation in Neural Marked Point Processes: Architectures and Loss Functions
• Deep Neyman-Scott Processes