Research map / Time series / Causal and generative time series
Safe Active Learning for Time-Series Modeling with Gaussian Processes
Christoph Zimmer, Mona Meister, Duy Nguyen-Tuong
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 SAL-NX information-gain trajectory selection instead of random selection under the same safety constraint | GP with NX structure, 4D ramps · Synthetic Experiment 2 | RMSE of model approximation | Improved |
| Use SAL-NX instead of random selection with safe constraints | GP with NX structure, 4D ramps · Synthetic Experiment 2 | coverage of safe regions | Improved |
| Use SAL-NX instead of random selection with safe constraints, 250 ramp trajectories, alpha=0.5 | High-pressure fluid injection surrogate model (7D input) · High-pressure fluid system | model approximation error and safe-region coverage | Improved |
| Vary safety threshold alpha over 4 values (ablation) | SAL-NX · Synthetic Experiment 2 | RMSE | Mixed |
| Vary safety threshold alpha (ablation) | SAL-NX · Synthetic Experiment 2 | percentage of unsafe trajectories | Mixed |
| Vary safety threshold alpha (ablation) | SAL-NX · High-pressure fluid system | approximation error and failures during exploration | Mixed |
| Replace Fisher-information criterion (adapted with GP mean) by the GP-variance determinant criterion, without safety | SAL-NX · Synthetic Experiment 2 (Appendix 7.4) | RMSE decrease | No clear effect |
| Add more exploration iterations (3 to 13 to 24) in the toy 2D example | SAL-NX · Synthetic Experiment 1 | RMSE and safe-region estimate | Improved |
Closest papers by meaning
- Safe Active Learning with Future Viability Guarantees in Time-Series Models (NeurIPS 2026 · similarity 0.89)
- Gaussian Process-Based Real-Time Learning for Safety Critical Applications (ICML 2021 · similarity 0.86)
- Active Learning for Derivative-Based Global Sensitivity Analysis with Gaussian Processes (NeurIPS 2024 · similarity 0.83)
- RobustTSF: Towards Theory and Design of Robust Time Series Forecasting with Anomalies (ICLR 2024 · similarity 0.83)
- Safely Learning Controlled Stochastic Dynamics (NeurIPS 2025 · similarity 0.83)
- TimeES: Probabilistic and Deterministic Time Series Forecasting via Evolutionary Spectra (NeurIPS 2026 · similarity 0.82)
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.
