Intensity-Free Learning of Temporal Point Processes
This work addresses a fundamental modeling challenge in event sequence analysis, offering a novel approach that enables new applications like sequence embeddings and data imputation.
The paper tackles the limitations of intensity-based learning in temporal point processes by directly modeling the conditional distribution of inter-event times using normalizing flows and a mixture model, achieving state-of-the-art performance in prediction tasks.
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.