Research map / Learning theory and algorithms / Prediction and decision losses
StructED: Risk Minimization in Structured Prediction
Yossi Adi, Joseph Keshet
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 |
|---|---|---|---|
| use orbit loss instead of SSVM | structured prediction training · Vowel duration measurement (test) | task loss (msec): 72.0 -> 58.5 | Improved |
| use direct loss minimization instead of SSVM | structured prediction training · Vowel duration measurement (test) | task loss (msec): 72.0 -> 63.5 | Improved |
| use structured ramp loss instead of SSVM | structured prediction training · Vowel duration measurement (test) | task loss (msec): 72.0 -> 64.5 | Improved |
| use structured probit loss instead of SSVM | structured prediction training · Vowel duration measurement (test) | task loss (msec): 72.0 -> 64.5 | Improved |
| use CRF instead of SSVM | structured prediction training · Vowel duration measurement (test) | task loss (msec): 72.0 -> 70.5 | No clear effect |
| use StructED SSVM instead of PyStruct SSVM | SSVM with SGD · MNIST | accuracy (%): 90.2 -> 92.6 | No clear effect |
Closest papers by meaning
- Stochastic Structured Prediction under Bandit Feedback (NeurIPS 2016 · similarity 0.84)
- Localized Structured Prediction (NeurIPS 2019 · similarity 0.83)
- A Smoother Way to Train Structured Prediction Models (NeurIPS 2018 · similarity 0.83)
- Resource Constrained Structured Prediction (AAAI 2017 · similarity 0.83)
- Structured Prediction Theory Based on Factor Graph Complexity (NeurIPS 2016 · similarity 0.82)
- Bandit and Delayed Feedback in Online Structured Prediction (NeurIPS 2025 · similarity 0.80)
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.
