Research map / Learning theory and algorithms / Prediction and decision losses
Neural Optimal Transport Meets Multivariate Conformal Prediction
Vladimir Kondratyev, Alexander Fishkov, Mahmoud Hegazy, Nikita Kotelevskii, Rémi Flamary, Maxim Panov, Eric Moulines
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 |
|---|---|---|---|
| amortize the conjugate optimization with a learned predictor (ablation) | C-NQR (L-BFGS conjugate) · synthetic datasets | training time per epoch: 15.08 sec. -> 8.89 sec. | Improved |
| amortize the conjugate optimization (ablation) | C-NQR · synthetic datasets | inference time: 1.21 sec. -> 1.12 sec. | Improved |
| use entropic regularization instead of exact conjugates (ablation) | NQR · Banana | S-W2: 0.072 -> 0.111 | Hurt |
| use entropic regularization instead of exact conjugates (ablation) | NQR · Glasses | S-W2: 0.785 -> 0.748 | Improved |
| estimate potential psi(y,x) instead of phi(u,x) (ablation) | AC-NQR · Star | S-W2: 0.182 -> 0.197 | Hurt |
| use neural conditional OT instead of discrete VQR or FN-VQR | vector quantile regression · Glasses | S-W2: 1.964 -> 0.771 | Improved |
| use neural conditional OT instead of CPF | vector quantile regression · Banana | S-W2: 0.069 -> 0.072 | No clear effect |
| add re-ranking step to pullback conformal sets (RPB vs PB) (ablation) | AC-NQR_U conformal · multi-target regression (scm20d, sgemm, blog, bio) | conditional coverage vs volume | Mixed |
| use pullback conformal sets on OT ranks | AC-NQR_U · multi-target regression | prediction set volume | Improved |
Closest papers by meaning
- Optimal transport-based conformal prediction (ICML 2025 · similarity 0.92)
- Co-optimization for Adaptive Conformal Prediction (NeurIPS 2026 · similarity 0.89)
- Rectifying Conformity Scores for Better Conditional Coverage (ICML 2025 · similarity 0.87)
- Robust Conditional Conformal Prediction via Branched Normalizing Flow (NeurIPS 2026 · similarity 0.87)
- Conformal Prediction with Learned Features (ICML 2024 · similarity 0.87)
- Non-exchangeable Conformal Prediction with Optimal Transport: Tackling Distribution Shift with Unlabeled Data (NeurIPS 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.
