Research map / Label-efficient and robust learning / Semi-supervised learning
Human Guided Linear Regression With Feature-Level Constraints
Aubrey Gress, Ian Davidson
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 |
|---|---|---|---|
| Add soft-relaxed sign guidance (PSRC) to Ridge | Ridge regression · Synthetic | test error: 0.179(0.029) -> 0.141(0.026) | Improved |
| Add soft-relaxed sign guidance (PSRC) to Ridge | Ridge regression · Boston Housing (BH) | test error: 0.202(0.036) -> 0.149(0.018) | Improved |
| Add soft-relaxed sign guidance (PSRC) to Ridge | Ridge regression · Concrete | test error: 0.275(0.020) -> 0.220(0.017) | Improved |
| Add soft-relaxed sign guidance (PSRC) to Ridge | Ridge regression · Heart | test error: 2.159(0.220) -> 2.007(0.191) | Improved |
| Use relaxed sign constraints instead of hard constraints (Signed Ridge) (ablation) | PSRC · ITS | test error: 0.586(0.095) -> 0.525(0.089) | Improved |
| Use relaxed sign constraints instead of hard constraints (Signed Ridge) (ablation) | PSRC · Heart | test error: 2.128(0.220) -> 2.007(0.191) | Improved |
| Use relaxed sign constraints instead of hard constraints (Signed Ridge) (ablation) | PSRC · Wine | test error: 0.094(0.012) -> 0.088(0.010) | Improved |
| Take sign guidance from outside source (simulated expert) rather than from training data (ablation) | PSRC · Concrete | test error: 0.266(0.020) -> 0.220(0.017) | Improved |
| Take sign guidance from outside source (simulated expert) rather than from training data (ablation) | PSRC · ITS | test error: 0.608(0.102) -> 0.525(0.089) | Improved |
| Use pairwise sign guidance (PPSCR) instead of sign guidance (PSRC) | PSRC · Housing | test error: 0.409(0.038) vs 0.399(0.037) | Mixed |
| Use relative ordering guidance (PRCR) instead of sign guidance (PSRC) | PSRC · Synthetic | test error: 0.141(0.026) -> 0.155(0.031) | Hurt |
| Transfer sign guidance from a source domain (Transfer PSRC) instead of Transfer Ridge | Transfer Ridge · BH | test error: 0.323(0.062) -> 0.242(0.045) | Improved |
Closest papers by meaning
- Covariate-Guided Clusterwise Linear Regression for Generalization to Unseen Data (ICLR 2026 · similarity 0.85)
- Semi-supervised Active Linear Regression (NeurIPS 2022 · similarity 0.85)
- Regression Learning with Limited Observations of Multivariate Outcomes and Features (ICML 2024 · similarity 0.84)
- Regression with Multi-Expert Deferral (ICML 2024 · similarity 0.84)
- Gradient Aligned Regression via Pairwise Losses (ICML 2025 · similarity 0.83)
- Transfer Learning of Linear Regression with Multiple Pretrained Models: Benefiting from More Pretrained Models via Overparameterization Debiasing (NeurIPS 2026 · 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.
