Research map / Probabilistic and scientific modeling / Causal discovery
Learning Hawkes Processes Under Synchronization Noise
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 |
|---|---|---|---|
| add synchronization-noise offset estimation with smoothed kernel and SGD (DESYNC-MHP MLE) instead of classic MLE | Multivariate Hawkes process edge recovery · Synthetic MHP, d=10, noise variance 1/beta | edge identification accuracy: near 100% vs. classic MLE misidentifying >25% of edges | Helped |
| add noise-aware estimation (DESYNC-MHP MLE) vs classic MLE | Multivariate Hawkes process on neuronal spike trains · Real neuronal spike data, predictive log-likelihood | test predictive log-likelihood: 0.4311 vs 0.4282 | Helped |
| increase number of dimensions d from 6 to 16 with proposed method (ablation) | DESYNC-MHP MLE · Synthetic MHP, sigma^2=1 and 5 | edge identification accuracy | No clear effect |
| increase number of independent realizations C for SGD (ablation) | DESYNC-MHP MLE · Synthetic MHP, d=10, sigma^2=1 | edge identification accuracy: C=3 sufficient to reach near 100% accuracy | Helped |
| add noise-aware estimation in high-noise regime (noise variance much larger than 1/beta) (ablation) | DESYNC-MHP MLE · Synthetic MHP, high noise | edge identification accuracy | No clear effect |
Closest papers by meaning
- Cumulants of Hawkes Processes are Robust to Observation Noise (ICML 2021 · similarity 0.94)
- Learning Hawkes Processes from a handful of events (NeurIPS 2019 · similarity 0.89)
- Causal Structure Learning in Hawkes Processes with Complex Latent Confounder Networks (ICLR 2026 · similarity 0.89)
- Uncertainty Quantification for Inferring Hawkes Networks (NeurIPS 2020 · similarity 0.89)
- Learning the Causal Structure of Networked Dynamical Systems under Latent Nodes and Structured Noise (AAAI 2024 · similarity 0.86)
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes (NeurIPS 2026 · similarity 0.85)
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.