Research map / Probabilistic and scientific modeling / Variational and generative sampling
Continuous-time edge modelling using non-parametric point processes
Xuhui Fan, Bin Li, Feng Zhou, Scott A Sisson
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 the NP-HP-C edge-clustering variant instead of NP-HP (ablation) | NP-HP · Email | AUC: 81.78 -> 85.29 | Improved |
| Use NP-HP-C instead of NP-HP (ablation) | NP-HP · Email | testing log-likelihood: −53 121 -> −50 126 | Improved |
| Use NP-HP-C instead of NP-HP (ablation) | NP-HP · Ubuntu | AUC: 86.78 -> 88.02 | Improved |
| Use NP-HP-C instead of NP-HP (ablation) | NP-HP · Overflow | testing log-likelihood: −114 277 -> −125 387 | Hurt |
| Use NP-HP-C instead of NP-HP (ablation) | NP-HP · College | AUC: 81.57 -> 82.17 | Mixed |
| Add excitation (Hawkes) to non-parametric Poisson model: NP-HP vs NP-PP (ablation) | NP-PP · Email | testing log-likelihood: −68 475 -> −53 121 | Improved |
| Add excitation (Hawkes) to NP-PP (ablation) | NP-PP · Overflow | AUC: 89.15 -> 89.58 | No clear effect |
| Non-parametric NP-HP vs parametric Hawkes-IRM baseline | Hawkes-IRM · Overflow | testing log-likelihood: −88 746 -> −114 277 | Hurt |
Closest papers by meaning
- Hawkes Processes With Stochastic Exogenous Effects for Continuous-Time Interaction Modelling (TPAMI 2022 · similarity 0.86)
- Efficient Inference for Nonparametric Hawkes Processes Using Auxiliary Latent Variables (JMLR 2020 · similarity 0.85)
- Proper Loss Functions for Nonlinear Hawkes Processes (AAAI 2018 · similarity 0.85)
- Isotonic Hawkes Processes (ICML 2016 · similarity 0.85)
- Scalable and Adaptive Variational Bayes Methods for Hawkes Processes (JMLR 2025 · similarity 0.84)
- Hyper Hawkes Processes: Interpretable Models of Marked Temporal Point Processes (NeurIPS 2026 · similarity 0.84)
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.
