Research map / Learning theory and algorithms / Prediction and decision losses
Systematic Bias of Machine Learning Regression Models and Correction
Hwiyoung Lee, Shuo Chen
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 two-sided mean constraints to LASSO (ablation) | LASSO · Synthetic linear, low noise (test) | Bias (slope): 0.1939 -> 0.0138 | Improved |
| add two-sided mean constraints to KRR (ablation) | KRR · Synthetic linear, low noise (test) | Bias (slope): 0.1147 -> 0.0350 | Improved |
| add constraints to LASSO (ablation) | LASSO · Synthetic linear, low noise (test) | RMSE: 0.5606 -> 0.5906 | Hurt |
| add constraints to KRR (ablation) | KRR · Synthetic linear, low noise (test) | RMSE: 0.5509 -> 0.5571 | Mixed |
| add constraints to KRR (ablation) | KRR · Synthetic nonlinear 2, low noise (test) | Bias (slope): 0.3369 -> 0.0274 | Improved |
| add constraints to KRR (ablation) | KRR · Synthetic linear, high noise (test) | Bias (slope): 0.3047 -> 0.1106 | Improved |
| add constraints to LASSO (ablation) | LASSO · Synthetic linear, low noise (test) | Cor(pred, residual): -0.5717 -> -0.0416 | Improved |
| add constraints to KRR (ablation) | KRR · Brain age HCP-A (test) | Bias (slope): 0.5014 -> 0.0836 | Improved |
| add constraints to LASSO (ablation) | LASSO · Brain age UKBB (test) | Bias (slope): 0.7046 -> 0.1423 | Improved |
| add constraints to KRR (ablation) | KRR · Brain age UKBB (test) | RMSE: 6.1274 -> 10.7556 | Hurt |
| add constraints to KRR (ablation) | KRR · Brain age HCP-A (test) | RMSE: 9.4402 -> 13.0586 | Hurt |
Closest papers by meaning
- From Hope to Safety: Unlearning Biases of Deep Models via Gradient Penalization in Latent Space (AAAI 2024 · similarity 0.84)
- Adaptive Shrinkage Estimation for Personalized Deep Kernel Regression in Modeling Brain Trajectories (ICLR 2025 · similarity 0.84)
- An Effective Theory of Bias Amplification (ICLR 2025 · similarity 0.84)
- Fast Model DeBias with Machine Unlearning (NeurIPS 2023 · similarity 0.83)
- Controlling for Omitted Variable Bias in Deep Neural Networks (NeurIPS 2026 · similarity 0.82)
- Mind the Gap: A Causal Perspective on Bias Amplification in Prediction & Decision-Making (NeurIPS 2024 · similarity 0.81)
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.
