# Automatic Integration for Spatiotemporal Neural Point Processes

arXiv: https://arxiv.org/abs/2310.06179 · HTML: https://mlsanity.com/paper/2310.06179/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 37 references that are in the map.
- First posted on arXiv 9 Oct 2023.

## Builds on (most cited first, 9 of 9)

- [AutoInt: Automatic Integration for Fast Neural Volume Rendering](https://mlsanity.com/paper/2012.01714/) (2020)
- [Intensity-Free Learning of Temporal Point Processes](https://mlsanity.com/paper/1909.12127/) (2019)
- [Fully Neural Network based Model for General Temporal Point Processes](https://mlsanity.com/paper/1905.09690/) (2019)
- [Neural Spatio-Temporal Point Processes](https://mlsanity.com/paper/2011.04583/) (2020)
- [Wasserstein Learning of Deep Generative Point Process Models](https://mlsanity.com/paper/1705.08051/) (2017)
- [Neural Point Process for Learning Spatiotemporal Event Dynamics](https://mlsanity.com/paper/2112.06351/) (2021)
- [A Review of Self-Exciting Spatio-Temporal Point Processes and Their Applications](https://mlsanity.com/paper/1708.02647/) (2017)
- [Learning Temporal Point Processes via Reinforcement Learning](https://mlsanity.com/paper/1811.05016/) (2018)
- [Deep Reinforcement Learning of Marked Temporal Point Processes](https://mlsanity.com/paper/1805.09360/) (2018)

## Built on by (most cited first, 13 of 13)

- [Unlocking Point Processes through Point Set Diffusion](https://mlsanity.com/paper/2410.22493/) (2024)
- [Neural Spatiotemporal Point Processes: Trends and Challenges](https://mlsanity.com/paper/2502.09341/) (2025)
- [Learning Spatiotemporal Dynamical Systems from Point Process Observations](https://mlsanity.com/paper/2406.00368/) (2024)
- [Back to Bayesics: Uncovering Human Mobility Distributions and Anomalies with an Integrated Statistical and Neural Framework](https://mlsanity.com/paper/2410.01011/) (2024)
- [EarthquakeNPP: A Benchmark for Earthquake Forecasting with Neural Point Processes](https://mlsanity.com/paper/2410.08226/) (2024)
- [Neural-ANOVA: Analytical Model Decomposition using Automatic Integration](https://mlsanity.com/paper/2408.12319/) (2024)
- [Deep spatio-temporal point processes: Advances and new directions](https://mlsanity.com/paper/2504.06364/) (2025)
- [Preventing Conflicting Gradients in Neural Marked Temporal Point Processes](https://mlsanity.com/paper/2412.08590/) (2024)
- [Addressing Mark Imbalance in Integration-free Neural Marked Temporal Point Processes](https://mlsanity.com/paper/2510.20414/) (2025)
- [Kronecker-Structured Nonparametric Spatiotemporal Point Processes](https://mlsanity.com/paper/2603.23746/) (2026)
- [Arbitrarily Conditioned Hierarchical Flows for Spatiotemporal Events](https://mlsanity.com/paper/2605.01226/) (2026)
- [Seahorse: A Unified Benchmarking Framework for Spatiotemporal Event Modeling](https://mlsanity.com/paper/2607.01022/) (2026)
- [From Jumps to Signatures: a Generative Method for Temporal Point Processes](https://mlsanity.com/paper/2607.06652/) (2026)

## Abstract (arXiv)

Learning continuous-time point processes is essential to many discrete event forecasting tasks. However, integration poses a major challenge, particularly for spatiotemporal point processes (STPPs), as it involves calculating the likelihood through triple integrals over space and time. Existing methods for integrating STPP either assume a parametric form of the intensity function, which lacks flexibility; or approximating the intensity with Monte Carlo sampling, which introduces numerical errors. Recent work by Omi et al. [2019] proposes a dual network approach for efficient integration of flexible intensity function. However, their method only focuses on the 1D temporal point process. In this paper, we introduce a novel paradigm: AutoSTPP (Automatic Integration for Spatiotemporal Neural Point Processes) that extends the dual network approach to 3D STPP. While previous work provides a foundation, its direct extension overly restricts the intensity function and leads to computational challenges. In response, we introduce a decomposable parametrization for the integral network using ProdNet. This approach, leveraging the product of simplified univariate graphs, effectively sidesteps the computational complexities inherent in multivariate computational graphs. We prove the consistency of AutoSTPP and validate it on synthetic data and benchmark real-world datasets. AutoSTPP shows a significant advantage in recovering complex intensity functions from irregular spatiotemporal events, particularly when the intensity is sharply localized. Our code is open-source at https://github.com/Rose-STL-Lab/AutoSTPP.