Research map / Learning theory and algorithms / Prediction and decision losses
A Consistent Regularization Approach for Structured Prediction
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 squared Hellinger loss instead of Gaussian loss for USPS reconstruction | proposed method (least squares surrogate) · USPS digit reconstruction | recognition loss | Helped |
| Replace KDE-style Gaussian loss with the proposed framework for ranking | proposed method on MovieLens ranking · MovieLens ranking | normalized rank loss | Helped |
| Use least squares surrogate with cauchy loss for robust regression instead of Nadaraya-Watson or KRR | proposed method on robust regression · sin regression with outliers | distance to true function | Helped |
Closest papers by meaning
- Leveraging Low-Rank Relations Between Surrogate Tasks in Structured Prediction (ICML 2019 · similarity 0.91)
- Structured Prediction with Stronger Consistency Guarantees (NeurIPS 2023 · similarity 0.90)
- On Structured Prediction Theory with Calibrated Convex Surrogate Losses (NeurIPS 2017 · similarity 0.89)
- Towards Sharper Generalization Bounds for Structured Prediction (NeurIPS 2021 · similarity 0.89)
- Bandit and Delayed Feedback in Online Structured Prediction (NeurIPS 2025 · similarity 0.86)
- Realizable $H$-Consistent and Bayes-Consistent Loss Functions for Learning to Defer (NeurIPS 2024 · similarity 0.86)
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.