Research map / Time series / Causal and generative time series
Neural Spectral Marked Point Processes
Shixiang Zhu, Haoyun Wang, Zheng Dong, Xiuyuan Cheng, Yao Xie
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 a neural finite-rank non-stationary influence kernel instead of the strongest listed baseline | NSMPP · 1D Synthetic 1, 1D Synthetic 2, 1D Synthetic 3, 2D Synthetic 1, 2D Synthetic 2, 2D Synthetic 3 | predictive log-likelihood: 1D Synthetic 1: -29.84 -> -17.68; 1D Synthetic 2: -36.10 -> -14.17; 1D Synthetic 3: -89.56 -> -21.94; 2D Synthetic 1: -2 | Improved |
| use a neural finite-rank non-stationary influence kernel instead of the strongest listed baseline | NSMPP · 1D Synthetic 1, 1D Synthetic 2, 1D Synthetic 3, 2D Synthetic 1, 2D Synthetic 2, 2D Synthetic 3 | MAE: 1D Synthetic 1: 2.65 -> 0.24; 1D Synthetic 2: 0.33 -> 0.02; 1D Synthetic 3: 0.25 -> 0.02; 2D Synthetic 1: 98.32 -> 10.16 | Improved |
| use a neural finite-rank non-stationary influence kernel instead of the strongest listed baseline | NSMPP · Earthquake (2D) and Robbery (1D) | predictive log-likelihood: Earthquake (2D): -189.39 -> -56.50; Robbery (1D): -96.10 -> -74.47 | Improved |
| increase network size by adding one hidden layer and doubling sub-network layer width (ablation) | NSMPP · 1D synthetic data | conditional-intensity prediction accuracy | No clear effect |
| train with half the default training samples (ablation) | NSMPP · 1D synthetic data | conditional-intensity prediction accuracy | No clear effect |
Closest papers by meaning
- Decoupled Marked Temporal Point Process using Neural Ordinary Differential Equations (ICLR 2024 · similarity 0.85)
- EasyTPP: Towards Open Benchmarking Temporal Point Processes (ICLR 2024 · similarity 0.85)
- Unlocking Point Processes through Point Set Diffusion (ICLR 2025 · similarity 0.85)
- Interacting Diffusion Processes for Event Sequence Forecasting (ICML 2024 · similarity 0.84)
- Decomposable Transformer Point Processes (NeurIPS 2024 · similarity 0.84)
- Addressing Mark Imbalance in Integration-free Marked Temporal Point Processes (NeurIPS 2025 · 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.
