Learning Structure-enhanced Temporal Point Processes with Gromov-Wasserstein Regularization

2025-03-29

AI Summary

Existing temporal point process (TPP) models often neglect the inherent clustering structure of event sequences, leading to limited interpretability. To address this, we propose a structure-enhanced TPP model that, for the first time, incorporates the Gromov–Wasserstein (GW) distance into the maximum likelihood estimation framework. Our method explicitly captures intrinsic clustering patterns via structured clustering constraints on sequence-level embeddings. It integrates nonparametric kernel functions, sequence similarity matrix construction, sampling-accelerated GW computation, and end-to-end joint optimization. Evaluated on multiple benchmark datasets, the model achieves state-of-the-art clustering quality and enhanced interpretability—evidenced by well-separated clusters in the embedding space—while preserving predictive accuracy. The core innovation lies in the deep coupling of GW regularization with TPP modeling, enabling synergistic improvement in both interpretability and predictive performance.

Technology Category

• Machine Learning: Structured Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Real-world event sequences are often generated by different temporal point processes (TPPs) and thus have clustering structures. Nonetheless, in the modeling and prediction of event sequences, most existing TPPs ignore the inherent clustering structures of the event sequences, leading to the models with unsatisfactory interpretability. In this study, we learn structure-enhanced TPPs with the help of Gromov-Wasserstein (GW) regularization, which imposes clustering structures on the sequence-level embeddings of the TPPs in the maximum likelihood estimation framework.In the training phase, the proposed method leverages a nonparametric TPP kernel to regularize the similarity matrix derived based on the sequence embeddings. In large-scale applications, we sample the kernel matrix and implement the regularization as a Gromov-Wasserstein (GW) discrepancy term, which achieves a trade-off between regularity and computational efficiency.The TPPs learned through this method result in clustered sequence embeddings and demonstrate competitive predictive and clustering performance, significantly improving the model interpretability without compromising prediction accuracy.

Problem

• Modeling clustered event sequences in temporal point processes
• Improving interpretability without losing prediction accuracy
• Balancing computational efficiency with structural regularization

Innovation

• Uses Gromov-Wasserstein regularization for clustering
• Incorporates nonparametric TPP kernel for similarity
• Balances computational efficiency with model regularity