Research map / Learning theory and algorithms / Prediction and decision losses
Maximum-Variance Total Variation Denoising for Interpretable Spatial Smoothing
Wesley Tansey, Jesse Thomason, James Scott
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 |
|---|---|---|---|
| MVTV vs CRISP | CRISP · Chicago crime data | AIC: 47245.57 -> 34016.60 | Improved |
| MVTV vs CRISP | CRISP · Austin crime data | AIC: 18326.33 -> 10327.59 | Improved |
| MVTV vs CART | CART · Chicago crime data | AIC: 43804.69 -> 34016.60 | Improved |
| MVTV vs CART | CART · Austin crime data | AIC: 11139.29 -> 10327.59 | Improved |
| TV smoothing over maximum-variance grid vs CRISP on same grid (MV-CRISP) (ablation) | MV-CRISP · Austin crime data | plateau count: 1361.75 -> 384.35 | Improved |
| TV smoothing over maximum-variance grid vs CRISP on same grid (MV-CRISP) (ablation) | MV-CRISP · Chicago crime data | AIC: 45314.71 -> 34016.60 | Improved |
| use maximum-variance q for CRISP (MV-CRISP vs CRISP) (ablation) | CRISP · Chicago crime data | plateau count: 9330.60 -> 8278.90 | Improved |
| use maximum-variance q for CRISP (MV-CRISP vs CRISP) (ablation) | CRISP · Austin crime data | AIC: 18326.33 -> 12064.25 | Improved |
| MVTV vs CRISP | CRISP · Chicago crime data | RMSE: 0.84 -> 0.86 | Hurt |
| MVTV vs CRISP | CRISP · Austin crime data | RMSE: 0.94 -> 0.97 | Hurt |
| MVTV vs CART | CART · Austin crime data | RMSE: 1.05 -> 0.97 | Improved |
| smooth data with MVTV instead of showing raw data | raw data · Austin crime data human study | human error x10^-2: 4.71±0.539 -> 2.75±0.334 | Improved |
Closest papers by meaning
- Ensembling Diffusion Models via Adaptive Feature Aggregation (ICLR 2025 · similarity 0.84)
- Spatial Reasoning with Denoising Models (ICML 2025 · similarity 0.83)
- Learning from Highly Sparse Spatio-temporal Data (NeurIPS 2024 · similarity 0.83)
- Reducing Spatial Fitting Error in Distillation of Denoising Diffusion Models (AAAI 2024 · similarity 0.83)
- Free Hunch: Denoiser Covariance Estimation for Diffusion Models Without Extra Costs (ICLR 2025 · similarity 0.83)
- Smooth Sailing: Lipschitz-Driven Uncertainty Quantification for Spatial Associations (NeurIPS 2025 · similarity 0.83)
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.
