Scalable Marked Point Processes for Exchangeable and Non-Exchangeable Event Sequences
This work addresses the problem of scalable event sequence modeling for researchers and practitioners, offering a more flexible and efficient alternative to existing methods, though it appears incremental as it builds on Hawkes processes and variational inference.
The authors tackled the challenge of scalable inference for marked point processes by developing a framework that handles both exchangeable and non-exchangeable event sequences without pre-training or extensive tuning, achieving competitive computational and predictive performance in real data experiments, including a case study on English Premier League events.
We adopt the interpretability offered by a parametric, Hawkes-process-inspired conditional probability mass function for the marks and apply variational inference techniques to derive a general and scalable inferential framework for marked point processes. The framework can handle both exchangeable and non-exchangeable event sequences with minimal tuning and without any pre-training. This contrasts with many parametric and non-parametric state-of-the-art methods that typically require pre-training and/or careful tuning, and can only handle exchangeable event sequences. The framework's competitive computational and predictive performance against other state-of-the-art methods are illustrated through real data experiments. Its attractiveness for large-scale applications is demonstrated through a case study involving all events occurring in an English Premier League season.