Research map / Learning theory and algorithms / Prediction and decision losses
Geometric Conformal Prediction with Spatial Ranks and Multivariate Quantiles
Anton Conrad, Eric Moulines, julien perez
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 geometric spatial rank (GRCP) instead of OT ranks for vector conformity scores | GRCP · Calibration score runtime, d=50 | wall-clock time: roughly 1.5× faster than OT-CP, 3.5× faster than OT-CP+ | Improved |
| use GRCP instead of OT-CP on high-cardinality classification | GRCP · EMNIST and Letter classification | average prediction-set size: OT-CP set size exceeds 15 | Improved |
| use GRCP instead of OT-CP/OT-CP+ in multi-output regression | GRCP · SCM1D and SCM20D regression | normalized log-volume | Improved |
| use GRCP instead of MultiDimSPCI/CopulaCP | GRCP · RF1, RF2, SGEMM regression | normalized log-volume | No clear effect |
| use smaller-region methods (GRCP, GCQR) instead of OT-CP+ / VSPS | GRCP/GCQR · RF1 and RF2 regression | worst-slab coverage | Hurt |
| use LAC scalar score instead of vector rank scores | LAC · Classification benchmarks | marginal coverage | Hurt |
| use RAPS instead of GRCP | RAPS · Classification benchmarks | set size and WSC | Mixed |
Closest papers by meaning
- Optimal transport-based conformal prediction (ICML 2025 · similarity 0.90)
- A Kernel Nonconformity Score for Multivariate Conformal Prediction (NeurIPS 2026 · similarity 0.88)
- Co-optimization for Adaptive Conformal Prediction (NeurIPS 2026 · similarity 0.87)
- Conformal Inference under High-Dimensional Covariate Shifts via Likelihood-Ratio Regularization (NeurIPS 2025 · similarity 0.87)
- Rectifying Conformity Scores for Better Conditional Coverage (ICML 2025 · similarity 0.87)
- A Unified Comparative Study with Generalized Conformity Scores for Multi-Output Conformal Regression (ICML 2025 · similarity 0.86)
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.
