Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Efficient Inference for Nonparametric Hawkes Processes Using Auxiliary Latent Variables
Feng Zhou, Zhidong Li, Xuhui Fan, Yang Wang, Arcot Sowmya, Fang Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use Gibbs sampler for nonparametric sigmoid Gaussian Hawkes instead of MLE with exponential kernel | SGHP · Simulated Case 3 | TestLL: 32.93 -> 38.94 (2.31) | Improved |
| use Gibbs sampler instead of MLE with exponential kernel | SGHP · Simulated Case 2 | TestLL: 27.92 -> 31.89 (1.52) | Improved |
| use Gibbs instead of MLE when ground truth is parametric | SGHP · Simulated Case 1 | TestLL: -55.75 -> -56.12 (2.05) | Hurt |
| allow time-varying baseline mu(t) (Gibbs vs WH with constant mu) | SGHP · Simulated Case 3 | EstErr(mu): 0.701 -> 0.165 (0.042) | Improved |
| use Gibbs instead of EM | SGHP · Simulated Case 3 | TestLL: 38.21 -> 38.94 (2.31) | Improved |
| use EM instead of Gibbs | SGHP · Simulated Case 3 | EstErr(mu): 0.165 (0.042) -> 0.134 | Improved |
| use nonparametric SGHP-Gibbs instead of MLE | SGHP · Crime in Vancouver | TestLL: 380.88 -> 430.43 (11.47) | Improved |
| use EM instead of Gibbs | SGHP · Crime in Vancouver | TestLL: 430.43 (11.47) -> 458.27 | Improved |
| use Hawkes self-exciting model (Gibbs) instead of Poisson SGCP | SGHP · Crime in Vancouver | TestLL: 299.77 (13.11) -> 430.43 (11.47) | Improved |
| use Hawkes self-exciting model (Gibbs) instead of Poisson SGCP | SGHP · NYPD complaints | TestLL: -292.30 (4.97) -> -193.76 (4.59) | Improved |
| add time-varying baseline (Gibbs vs WH constant mu) | SGHP · NYPD complaints | TestLL: -225.93 -> -193.76 (4.59) | Improved |
| use mean-field instead of Gibbs | SGHP · NYPD complaints | TestLL: -193.76 (4.59) -> -200.70 (3.32) | Hurt |
Closest papers by meaning
- Scalable and Adaptive Variational Bayes Methods for Hawkes Processes (JMLR 2025 · similarity 0.92)
- Variational Inference for Sparse Gaussian Process Modulated Hawkes Process (AAAI 2020 · similarity 0.92)
- Hawkes Process Inference With Missing Data (AAAI 2018 · similarity 0.90)
- Multivariate Hawkes Processes for Large-Scale Inference (AAAI 2017 · similarity 0.90)
- Learning Hawkes Processes from a handful of events (NeurIPS 2019 · similarity 0.89)
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
