Decomposable Transformer Point Processes

TL;DR

This paper introduces a novel Transformer-based framework for marked point processes that avoids computationally intensive inference algorithms, achieving state-of-the-art performance and faster predictions, especially for long-horizon tasks.

Contribution

It proposes a new method modeling inter-event times with a mixture of log-normals and marks with Transformers, eliminating the need for thinning during inference.

Findings

Achieves state-of-the-art next event prediction accuracy.

Outperforms thinning-based methods in inference speed.

Excels in long-horizon prediction tasks.

Abstract

The standard paradigm of modeling marked point processes is by parameterizing the intensity function using an attention-based (Transformer-style) architecture. Despite the flexibility of these methods, their inference is based on the computationally intensive thinning algorithm. In this work, we propose a framework where the advantages of the attention-based architecture are maintained and the limitation of the thinning algorithm is circumvented. The framework depends on modeling the conditional distribution of inter-event times with a mixture of log-normals satisfying a Markov property and the conditional probability mass function for the marks with a Transformer-based architecture. The proposed method attains state-of-the-art performance in predicting the next event of a sequence given its history. The experiments also reveal the efficacy of the methods that do not rely on the thinning algorithm during inference over the ones they do. Finally, we test our method on the challenging long-horizon prediction task and find that it outperforms a baseline developed specifically for tackling this task; importantly, inference requires just a fraction of time compared to the thinning-based baseline.