Research map / Learning theory and algorithms / Prediction and decision losses
On Regularization and Inference with Label Constraints
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 |
|---|---|---|---|
| Add regularization with constraints (violation penalty rho*V(f), semi-supervised) to the base model | multiclass classifiers with label constraints · theoretical analysis (no experiments) | generalization gap / optimal risk | Mixed |
| Apply constrained inference (CCM with mu=infinity, strictly constrained) post-training to a fixed model | base model with noise-free constraint · theoretical analysis (no experiments) | cross-entropy risk: Delta = V_ce(f) | Helped |
| Apply post-training constrained inference to a model whose expected violation is below the constraint noise rate | base model with noisy constraint · theoretical analysis (no experiments) | cross-entropy risk | Hurt |
| Use large regularization weight rho >= 1/(V_ora - V(f_inf)) before post-training constrained inference | regularized model with noisy constraint · theoretical analysis (no experiments) | cross-entropy risk | No clear effect |
| Combine on-training CCM with regularization (small rho) versus vanilla regularization | regularized CCM model · theoretical analysis (no experiments) | cross-entropy risk | Helped |
Closest papers by meaning
- Understanding the Impact of Introducing Constraints at Inference Time on Generalization Error (ICML 2024 · similarity 0.88)
- Understanding Domain Generalization: A Noise Robustness Perspective (ICLR 2024 · similarity 0.86)
- CGMatch: A Different Perspective of Semi-supervised Learning (CVPR 2025 · similarity 0.85)
- Data-Constrained Language Model Pretraining: Improved Regularization and Scaling Laws (NeurIPS 2026 · similarity 0.85)
- Learning from Noisy Labels via Conditional Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.85)
- Limits to scalable evaluation at the frontier: LLM as judge won’t beat twice the data (ICLR 2025 · 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.