Published on Thu Sep 26 2019

Intensity-Free Learning of Temporal Point Processes

Oleksandr Shchur, Marin Biloš, Stephan Günnemann

The standard way of learning in such models is by estimating the conditional intensity function. We show how to overcome the limitations of intensity-based approaches by directly modeling the conditional distribution of inter-event times. The proposed models achieve state-of-the-art performance in standard prediction tasks.

0
0
0
Abstract

Temporal point processes are the dominant paradigm for modeling sequences of events happening at irregular intervals. The standard way of learning in such models is by estimating the conditional intensity function. However, parameterizing the intensity function usually incurs several trade-offs. We show how to overcome the limitations of intensity-based approaches by directly modeling the conditional distribution of inter-event times. We draw on the literature on normalizing flows to design models that are flexible and efficient. We additionally propose a simple mixture model that matches the flexibility of flow-based models, but also permits sampling and computing moments in closed form. The proposed models achieve state-of-the-art performance in standard prediction tasks and are suitable for novel applications, such as learning sequence embeddings and imputing missing data.

Fri Oct 18 2019
Machine Learning
Point Process Flows
Event sequences can be modeled by temporal point processes (TPPs) to capture their asynchronous and probabilistic nature. We propose an intensity-free framework that directly models the point process distribution by utilizing normalizing flows. This approach is capable of capturing highly complex temporal distributions.
0
0
0
Sun Jun 20 2021
Machine Learning
Neural Spectral Marked Point Processes
0
0
0
Tue Jun 29 2021
Machine Learning
Continuous Latent Process Flows
Fitting this type of data using statistical models with continuous dynamics is promising at an intuitive level. It has practical benefits, including the ability to generate continuous journeys and to perform inference on previously unseen time stamps. We tackle these challenges with continuous latent process flows.
1
0
4
Tue Jul 28 2020
Machine Learning
UNIPoint: Universally Approximating Point Processes Intensities
Point processes are a useful mathematical tool for describing events over time. One notable open question is how to precisely describe the flexibility of point process models. We provide a proof that a class of learnable functions can universally approximate any valid intensity function.
0
0
0
Thu May 23 2019
Machine Learning
Fully Neural Network based Model for General Temporal Point Processes
A temporal point process is a mathematical model for a time series of discrete events. Recently, recurrent neural network (RNN) based models have been developed for point processes. We propose a novel RNN based model in which the time course of the intensity function is represented in a general manner.
0
0
0
Wed May 29 2019
Artificial Intelligence
Reinforcement Learning with Policy Mixture Model for Temporal Point Processes Clustering
Temporal point process is an expressive tool for modeling event sequences. In this paper, we take a reinforcement learning view whereby the observed sequences are assumed to be generated from a mixture of latent policies. The purpose is to cluster the sequences with different temporal patterns into the underlying policies.
0
0
0