Research map / Time series / Causal and generative time series
Adaptive Conformal Predictions for Time Series
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 |
|---|---|---|---|
| Increase ACI learning rate gamma from 0 to 0.01 or 0.05 on AR(1) residuals (ablation) | ACI · AR(1) toy example | empirical coverage: 72.6% -> 90% / 92% | Helped |
| Theory: increasing gamma in exchangeable case | ACI · exchangeable scores (Theorem 1) | expected interval length: gamma=0.03 -> +1.59% length; gamma=0.05 -> +3.38% (alpha=0.01, Gaussian) | Hurt |
| Adaptive aggregation of K ACI experts (AgACI) instead of fixed gamma | ACI · Synthetic ARMA(1,1) noise, T1=100 | coverage and efficiency | Helped |
| Naive strategy of picking past-best gamma instead of AgACI (ablation) | ACI gamma selection · Synthetic ARMA(1,1) noise | validity | Hurt |
| Use mean aggregation for EnbPI (EnbPI V2) instead of mixing mean and quantile aggregation | EnbPI · Synthetic ARMA(1,1) noise | coverage | Helped |
| Refit the regression model online (OSSCP) instead of offline SCP | Split conformal prediction · Synthetic AR(1)/ARMA(1,1) noise | coverage/length | Helped |
| Online refit of OSSCP on electricity prices | Split conformal prediction · French electricity spot prices 2019 | median length vs coverage | Helped |
| ACI with gamma 0.01 or 0.05 on day-ahead French electricity prices | ACI · French electricity spot prices 2019 | median length at correct coverage | Helped |
| AgACI aggregation on electricity prices | AgACI · French electricity spot prices 2019 | median length | Mixed |
Closest papers by meaning
- Error-quantified Conformal Inference for Time Series (ICLR 2025 · similarity 0.89)
- Conformal Prediction for Time-series Forecasting with Change Points (NeurIPS 2025 · similarity 0.89)
- Neural Conformal Control for Time Series Forecasting (AAAI 2025 · similarity 0.89)
- Kernel-based Optimally Weighted Conformal Time-Series Prediction (ICLR 2025 · similarity 0.86)
- Conformal Validity Guarantees Exist for Any Data Distribution (and How to Find Them) (ICML 2024 · similarity 0.86)
- Online Localized Conformal Prediction (NeurIPS 2026 · 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.