Research map / Learning theory and algorithms / Prediction and decision losses
Conformal Inference for Online Prediction with Arbitrary Distribution Shifts
Isaac Gibbs, Emmanuel J. Candès
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 DtACI (aggregating multiple step sizes) instead of MVP | online conformal prediction · Simulation, smooth and jump shifts | local coverage gap | Improved |
| use DtACI instead of AgACI under sharp jump shifts | online conformal prediction · Simulation, jump shift | coverage gap | Improved |
| use MVP or AgACI instead of DtACI in the stationary setting | online conformal prediction · Simulation, stationary | coverage gap | Hurt |
| use DtACI instead of AgACI on smooth shifts | online conformal prediction · Simulation, smooth shift | coverage | No clear effect |
| use DtACI instead of fixed alpha baseline | GARCH volatility prediction on four stocks · Stock volatility, target 0.9, normalized and unnormalized scores | local coverage | Improved |
| use DtACI instead of AgACI on stock data | GARCH volatility prediction · Stock volatility | local coverage | No clear effect |
| use DtACI instead of MVP on COVID-19 county forecasts | county case count prediction · COVID-19 county forecasts | local coverage | Improved |
| use variable eta_t instead of fixed heuristic eta (ablation) | DtACI · Simulation and stock experiments | coverage | No clear effect |
Closest papers by meaning
- Adaptive Conformal Inference by Betting (ICML 2024 · similarity 0.88)
- Error-quantified Conformal Inference for Time Series (ICLR 2025 · similarity 0.87)
- Online Conformal Prediction via Online Optimization (ICML 2025 · similarity 0.87)
- Conformal Validity Guarantees Exist for Any Data Distribution (and How to Find Them) (ICML 2024 · similarity 0.87)
- Online Localized Conformal Prediction (NeurIPS 2026 · similarity 0.86)
- Conformal Prediction for Time-series Forecasting with Change Points (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.
