Research map / Learning theory and algorithms / Sample complexity
Interpretable Predictive Modeling for Climate Variables with Weighted Lasso
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 distance-proportional weights in l1 penalty instead of uniform weights (Lasso) | Lasso · CMCC-CESM Brazil | RMSE: 0.6681 ± 0.0317 -> 0.6580 ± 0.0344 | No clear effect |
| use distance-proportional weights instead of uniform weights | Lasso · Brazil variable selection | selection stability and proximity | Helped |
| use weighted Lasso instead of Deep nets | Deep nets · INM-CM4 Brazil | RMSE: 0.9144 ± 0.0304 -> 0.7641 ± 0.0173 | Helped |
| use weighted Lasso instead of GBT | GBT · INM-CM4 Peru | RMSE: 0.8164 ± 0.0184 -> 0.7127 ± 0.0144 | Helped |
| use weighted Lasso instead of PCR | PCR · INM-CM4 Brazil | RMSE: 1.2030 ± 0.0637 -> 0.7641 ± 0.0173 | Helped |
| add l1 regularization to Deep nets (ablation) | Deep nets · CMCC-CESM Brazil | RMSE: 0.8151 ± 0.0364 -> 0.6931 ± 0.0260 | Helped |
| add l2 regularization to Deep nets (ablation) | Deep nets · CMCC-CESM Brazil | RMSE: 0.8151 ± 0.0364 -> 0.8458 ± 0.0744 | No clear effect |
| add l1+l2 regularization to Deep nets (ablation) | Deep nets · CMCC-CESM Brazil | RMSE: 0.8151 ± 0.0364 -> 1.0635 ± 0.1308 | Hurt |
| increase mini-batch size (ablation) | Deep nets (8 layers) · CMCC-CESM land targets | RMSE | Helped |
| increase hidden units from 1 to 64 (ablation) | Deep nets · CMCC-CESM Brazil | RMSE | No clear effect |
| use 8 hidden layers instead of 1 (ablation) | Deep nets · CMCC-CESM Brazil | RMSE | Helped |
| use pretrained Resnet-50 with transfer learning instead of MLP Deep nets (ablation) | Deep nets · CMCC-CESM Brazil | RMSE: 0.8151 ± 0.0364 -> 1.2972 ± 0.4109 | Hurt |
Closest papers by meaning
- Lazy Estimation of Variable Importance for Large Neural Networks (ICML 2022 · similarity 0.87)
- Common Task Framework For a Critical Evaluation of Scientific Machine Learning Algorithms (NeurIPS 2025 · similarity 0.85)
- DroughtSet: Understanding Drought Through Spatial-Temporal Learning (AAAI 2025 · similarity 0.83)
- A General Adaptive Dual-level Weighting Mechanism for Remote Sensing Pansharpening (CVPR 2025 · similarity 0.83)
- Multi-Scale Graph Learning for Anti-Sparse Downscaling (AAAI 2025 · similarity 0.83)
- Validating Climate Models with Spherical Convolutional Wasserstein Distance (NeurIPS 2024 · 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.