Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Learning Hawkes Processes from a handful of events
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| replace MLE-SGLP with variational EM (VI-SG) with per-parameter hyper-parameters | non-parametric Hawkes · Epidemics | predictive log-likelihood: -3.03 -> -2.06 | Helped |
| replace MLE-SGLP with VI-SG | non-parametric Hawkes · Stock market | predictive log-likelihood: -2.45 -> -1.00 | Helped |
| replace MLE-SGLP with VI-SG | non-parametric Hawkes · Enron email | predictive log-likelihood: -1.01 -> -0.42 | Helped |
| replace MLE-ADM4 with VI-EXP | parametric Hawkes · Epidemics | predictive log-likelihood: -4.61 -> -4.31 | Helped |
| replace MLE-ADM4 with VI-EXP | parametric Hawkes · Stock market | predictive log-likelihood: -2.81 -> -2.82 | No clear effect |
| replace MLE-ADM4 with VI-EXP | parametric Hawkes · Enron email | predictive log-likelihood: -0.40 -> -0.23 | Helped |
| use non-parametric excitation (Gaussian kernels) instead of exponential (ablation) | Hawkes learning · Epidemics and Stock market | predictive log-likelihood: -4.31 -> -2.06 (Epidemics VI) | Helped |
| use VI-SG instead of MLE-SGLP on short synthetic sequences | non-parametric Hawkes · Synthetic MHP N=5000 | F1-score: about 20% | Helped |
| increase number of basis functions M (ablation) | MLE-SGLP versus VI-SG · Synthetic N=2000 | robustness of F1/precision | Helped |
Closest papers by meaning
- Learning Hawkes Processes Under Synchronization Noise (ICML 2019 · similarity 0.89)
- Multivariate Hawkes Processes for Large-Scale Inference (AAAI 2017 · similarity 0.89)
- Causal Structure Learning in Hawkes Processes with Complex Latent Confounder Networks (ICLR 2026 · similarity 0.88)
- Learning Granger Causality for Hawkes Processes (ICML 2016 · similarity 0.88)
- Hawkes Process Inference With Missing Data (AAAI 2018 · similarity 0.88)
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes (NeurIPS 2026 · similarity 0.86)
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.