Research map / Time series / Causal and generative time series
Spatio-temporal point processes with deep non-stationary kernels
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 displacement-based low-rank kernel parameterization | DNSK+Barrier · 1D Data set 2 | intensity MRE: 0.016 vs 0.162 | Improved |
| use displacement-based low-rank kernel parameterization | DNSK+Barrier · 1D Data set 2 | testing log-likelihood per event: -2.536 vs -2.638 | Improved |
| use log-barrier instead of Softplus activation (ablation) | DNSK · Financial Transactions | testing log-likelihood: -0.889 -> --0.709 | Improved |
| use log-barrier instead of Softplus activation (ablation) | DNSK · Southern California earthquake data | time MAE: 1.685 -> 1.474 | Improved |
| use log-barrier instead of Softplus activation (ablation) | DNSK · 3D Data set 1 | intensity MRE: 0.088 -> 0.021 | Improved |
| use log-barrier instead of Softplus activation (ablation) | DNSK · StackOverflow | testing log-likelihood: -2.173 -> -2.089 | Improved |
| use DNSK+Barrier rather than NSMPP | spatio-temporal point-process model · 1D Data set 1 training | training time per epoch: 60.040 -> 1.299 | Improved |
| use DNSK+Barrier rather than NSMPP | spatio-temporal point-process model · 3D Data set 1 training | training time per epoch: 170.004 -> 3.529 | Improved |
| increase temporal and spatial grid resolution (ablation) | DNSK+Barrier · 3D Data set 2 | testing log-likelihood per event: -2.251_{(0.001)} to -2.250_{(0.001)} | No clear effect |
Closest papers by meaning
- Neural Spectral Marked Point Processes (ICLR 2022 · similarity 0.92)
- Integration-free Training for Spatio-temporal Multimodal Covariate Deep Kernel Point Processes (NeurIPS 2023 · similarity 0.88)
- Unlocking Point Processes through Point Set Diffusion (ICLR 2025 · similarity 0.85)
- Kernel Representation Learning for Time Sequence: Algorithm, Theory, and Applications (AAAI 2025 · similarity 0.85)
- Interacting Diffusion Processes for Event Sequence Forecasting (ICML 2024 · similarity 0.84)
- EasyTPP: Towards Open Benchmarking Temporal Point Processes (ICLR 2024 · 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.
