Structured Neural Marked Point Processes for Interpretable Event Interaction Modeling
For practitioners analyzing multi-class event streams, this work provides a model that balances high predictive accuracy with interpretable discovery of inter-event relationships, addressing a key limitation of black-box neural point processes.
The paper introduces a structured neural marked point process (SNMPP) that models multi-class event streams with an interpretable interaction network, enabling explicit discovery of excitation/inhibition relationships while achieving strong predictive performance on synthetic and real-world benchmarks.
Multi-class event streams arise in numerous real-world applications, where uncovering structured, interpretable inter-event relationships, together with accurate prediction, remains a central challenge. Existing neural point process models are highly expressive but encode event interactions in a black-box manner, preventing explicit discovery of structured dependencies. In this paper, we propose a structured neural marked point process (SNMPP) that achieves high modeling flexibility while enabling explicit event-wise and class-wise relationship discovery from data. Our model constructs a product-form neural influence kernel composed of a signed interaction network over event types and a delay-aware monotonic temporal network. This design enables explicit characterization of inter-class influence topology -- including excitation, inhibition, and neutrality -- while flexibly capturing diverse temporal decay patterns and potential influence delays. For efficient learning, we develop a stratified Monte Carlo estimator for stochastic training. Extensive experiments on synthetic and real-world benchmark datasets validate the ability of our approach to uncover structured relationships and deliver strong predictive performance.