Research map / Learning theory and algorithms / Sample complexity
Preference Modeling with Context-Dependent Salient Features
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 |
|---|---|---|---|
| Model comparisons using context-dependent salient features rather than a fixed feature set | District compactness ranking · District compactness test sets | Kendall tau correlation: outperforms both the MLE of FBTL and Ranking SVM by a significant amount on 6 out 7 test sets | Mixed |
| Use context-dependent salient features instead of FBTL's fixed full feature set | UT Zappos50K pairwise prediction · UT Zappos50K | pairwise prediction accuracy | No clear effect |
| Set top-t to include all features, yielding FBTL (ablation) | Synthetic top-t salient feature model · Synthetic pairwise preferences | intransitivity and pairwise inconsistencies | Hurt |
| Fit FBTL to data generated by the top-1 salient feature model | Synthetic preference data · Synthetic top-1 comparisons | estimation error of w* | Hurt |
Closest papers by meaning
- Active preference learning for ordering items in- and out-of-sample (NeurIPS 2024 · similarity 0.88)
- Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective (ICML 2025 · similarity 0.87)
- Generalizing while preserving monotonicity in comparison-based preference learning models (NeurIPS 2025 · similarity 0.87)
- Preference learning made easy: Everything should be understood through win rate (ICML 2025 · similarity 0.87)
- Preference Learning Algorithms Do Not Learn Preference Rankings (NeurIPS 2024 · similarity 0.86)
- DeepHalo: A Neural Choice Model with Controllable Context Effects (NeurIPS 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.