Research map / Probabilistic and scientific modeling / Differential equations and PDEs
Neural Jump Stochastic Differential Equations
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 |
|---|---|---|---|
| Use Neural JSDE instead of the fitted Poisson baseline | Conditional intensity modeling · Poisson | mean absolute percentage error: 0.1 -> 1.3 | Hurt |
| Use Neural JSDE instead of the fitted Hawkes exponential baseline | Conditional intensity modeling · Hawkes (E) | mean absolute percentage error: 3.5 -> 5.9 | Hurt |
| Use Neural JSDE instead of the fitted Hawkes power-law baseline | Conditional intensity modeling · Hawkes (PL) | mean absolute percentage error: 9.8 -> 17.1 | Hurt |
| Use Neural JSDE instead of the fitted self-correcting baseline | Conditional intensity modeling · Self-Correcting | mean absolute percentage error: 1.6 -> 9.3 | Hurt |
| Use Neural JSDE instead of the RNN baseline | Conditional intensity modeling · Poisson | mean absolute percentage error: 3.2 -> 1.3 | Helped |
| Use Neural JSDE instead of the RNN baseline | Conditional intensity modeling · Hawkes (E) | mean absolute percentage error: 22.0 -> 5.9 | Helped |
| Use Neural JSDE instead of the RNN baseline | Conditional intensity modeling · Hawkes (PL) | mean absolute percentage error: 20.1 -> 17.1 | Helped |
| Use Neural JSDE instead of the RNN baseline | Conditional intensity modeling · Self-Correcting | mean absolute percentage error: 24.3 -> 9.3 | Helped |
| Use Neural JSDE instead of prior event-type models | Discrete event type prediction · Stack Overflow | classification error rate: 53.7 -> 52.7 | Helped |
| Use Neural JSDE instead of the best reported prior model | Discrete event type prediction · MIMIC2 | classification error rate: 16.8 -> 19.8 | Hurt |
| Use Neural JSDE instead of predicting the running mean feature | Synthetic Hawkes event-feature prediction · Synthetic Hawkes event-feature prediction | mean absolute error: 3.654 -> 0.353 | Helped |
Closest papers by meaning
- Neural Jump-Diffusion Temporal Point Processes (ICML 2024 · similarity 0.89)
- Neural Stochastic Flows: Solver-Free Modelling and Inference for SDE Solutions (NeurIPS 2025 · similarity 0.88)
- Stable Neural Stochastic Differential Equations in Analyzing Irregular Time Series Data (ICLR 2024 · similarity 0.87)
- Decoupled Marked Temporal Point Process using Neural Ordinary Differential Equations (ICLR 2024 · similarity 0.85)
- Neural structure learning with stochastic differential equations (ICLR 2024 · similarity 0.85)
- Neural MJD: Neural Non-Stationary Merton Jump Diffusion for Time Series Prediction (NeurIPS 2025 · 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.