Research map / Learning theory and algorithms / Kernels and regression theory
Regularization Penalty Optimization for Addressing Data Quality Variance in OoD Algorithms
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 optimized environment-level IRM penalties and optimized sample-level Lipschitz penalties rather than uniform penalties | LipIRM/RPO · Cigar test | MSE: 6.54 ± 3.22 -> 4.35 ± 0.33 | Helped |
| Use optimized environment-level IRM penalties and optimized sample-level Lipschitz penalties rather than uniform penalties | LipIRM/RPO · Wage test | MSE: 34.93 ± 7.12 -> 2.82 ± 1.43 | Helped |
| Add optimized environment-level IRM penalties (ablation) | RPO · Colored-MNIST Setting 1 (Bad Environment) | accuracy: 0.398±1e-2 -> 0.643±4e-2 | Helped |
| Add optimized sample-level Lipschitz penalties (ablation) | RPO · Colored-MNIST Setting 1 (Bad Environment) | accuracy: 0.616±5e-2 -> 0.643±4e-2 | Helped |
| Add optimized environment-level IRM penalties (ablation) | RPO · Colored-MNIST Setting 7 (Bad Group) | accuracy: 0.679±3e-2 -> 0.699±8e-2 | Helped |
| Add optimized sample-level Lipschitz penalties (ablation) | RPO · Colored-MNIST Setting 7 (Bad Group) | accuracy: 0.544±1e-1 -> 0.699±8e-2 | Helped |
| Add optimized environment-level IRM penalties (ablation) | RPO · Colored-MNIST Setting 13 (Mixup) | accuracy: 0.431±2e-2 -> 0.595±3e-2 | Helped |
| Add optimized sample-level Lipschitz penalties (ablation) | RPO · Colored-MNIST Setting 13 (Mixup) | accuracy: 0.509±3e-2 -> 0.595±3e-2 | Helped |
Closest papers by meaning
- Pareto Invariant Risk Minimization: Towards Mitigating the Optimization Dilemma in Out-of-Distribution Generalization (ICLR 2023 · similarity 0.90)
- Generalization Bounds for Out-of-distribution Generalization (ICML 2026 · similarity 0.89)
- Out-of-distribution Generalization for Total Variation based Invariant Risk Minimization (ICLR 2025 · similarity 0.87)
- Towards Robust Out-of-Distribution Generalization Bounds via Sharpness (ICLR 2024 · similarity 0.87)
- Optimal Ridge Regularization for Out-of-Distribution Prediction (ICML 2024 · similarity 0.87)
- Robustly Train Normalizing Flows via KL Divergence Regularization (AAAI 2024 · similarity 0.85)
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.