Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
DistMatch: Adaptive Binning via Distribution Matching for Robust Sequential Conformal Prediction
Enver Menadjiev, Jihyeon Seong, Jisu Yeo, Jaesik Choi
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 |
|---|---|---|---|
| replace reweighting with KS-based binning of residuals into a matching tree | sequential conformal prediction (RF predictor) · Elec. Winkler score | Winkler score: 1.97 vs 2.54 (SPCI) | Improved |
| use KS statistic instead of KL as the distribution matching criterion (ablation) | DistMatch · Elec., Solar, Wind | Winkler score / coverage | Improved |
| use KS statistic instead of MI or Wasserstein distance as matching criterion (ablation) | DistMatch · Elec., Solar, Wind | computation time | Improved |
| use KS statistic as matching criterion versus alternatives (KL, MI, WD) (ablation) | DistMatch · Elec., Solar, Wind (average) | Winkler score | Improved |
| vary KS threshold gamma within 0.01 to 0.1 (ablation) | DistMatch · Elec., Solar, Wind | Winkler score | No clear effect |
| vary patch length w from 5 to 100 (ablation) | DistMatch · Elec., Wind (robust), Solar | Winkler score | Mixed |
| increase patch length from w=5 to w=10 (ablation) | DistMatch · patch-to-target smoothness ratio C | ratio C | Improved |
| retrain DistMatch online instead of fixed tree (ablation) | DistMatch · online deployment | performance | No clear effect |
| set nmin = 0 for stopping criterion (ablation) | DistMatch · leaf size | average leaf size: average leaf size above 10 | No clear effect |
| use DistMatch with LSTM backbone against probabilistic forecasting baselines (MDN, Gauss, MCD) | LSTM predictor · Elec., Solar, Wind | average coverage: 0.90 vs 0.67 (MDN) | Improved |
| use DistMatch instead of MDN on Solar with LSTM | LSTM predictor · Solar | Winkler score: 1.87 vs 1.53 (MDN, 1.53±0.25) | Hurt |
| use DistMatch instead of SPCI on Solar and Wind width (RF) | RF predictor · Solar width | Width: 60.00 vs 47.36 (SPCI) | Hurt |
Closest papers by meaning
- Conformal Prediction under Lévy-Prokhorov Distribution Shifts: Robustness to Local and Global Perturbations (NeurIPS 2025 · similarity 0.88)
- Co-optimization for Adaptive Conformal Prediction (NeurIPS 2026 · similarity 0.88)
- Kernel-based Optimally Weighted Conformal Time-Series Prediction (ICLR 2025 · similarity 0.88)
- Error-quantified Conformal Inference for Time Series (ICLR 2025 · similarity 0.88)
- Leave-One-Out Stable Conformal Prediction (ICLR 2025 · similarity 0.87)
- Non-exchangeable Conformal Prediction with Optimal Transport: Tackling Distribution Shift with Unlabeled Data (NeurIPS 2025 · similarity 0.87)
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.
