Research map / Learning theory and algorithms / Prediction and decision losses
Interpretable Preference Learning: A Game Theoretic Framework for Large Margin On-Line Feature and Rule Learning
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 Level 2 aggregated suit and rank features to Level 1 (ablation) | PRL · poker TOK vs rest | BACC (%): 48.08 -> 100.00 | Helped |
| Add Level 2 aggregated suit and rank features to Level 1 (ablation) | PRL · poker Flush vs rest | BACC (%): 59.03 -> 77.75 | Helped |
| Add Level 2 aggregated suit and rank features to Level 1 (ablation) | PRL · poker Straight vs rest | BACC (%): 49.97 -> 51.29 | No clear effect |
| Add Level 3 aggregate features to Level 2 (ablation) | PRL · poker TOK vs rest | BACC (%): 100.00 -> 100.00 | No clear effect |
| Add Level 3 aggregate features to Level 2 (ablation) | PRL · poker Flush vs rest | BACC (%): 77.75 -> 100.00 | Helped |
| Add Level 3 aggregate features to Level 2 (ablation) | PRL · poker Straight vs rest | BACC (%): 51.29 -> 100.00 | Helped |
| Use 10 most relevant rules rather than only the 4th-ranked rule (ablation) | PRL · breast-cancer | Accuracy (%): 92.67 -> 97.95 | Helped |
| Use 10 most relevant rules rather than 5 rules (ablation) | PRL · breast-cancer | Accuracy (%): 96.12 -> 97.95 | Helped |
| Use PRL instead of SVM | poker TOK vs rest, Level 3 · poker TOK vs rest, Level 3 | BACC (%): 99.99 -> 100.00 | No clear effect |
| Use PRL instead of SVM | poker Flush vs rest, Level 3 · poker Flush vs rest, Level 3 | BACC (%): 96.43 -> 100.00 | Helped |
| Use PRL instead of SVM | poker Straight vs rest, Level 3 · poker Straight vs rest, Level 3 | BACC (%): 50.00 -> 100.00 | Helped |
| Use PRL instead of SVM | dorothea · dorothea | Accuracy (%): 91.88 -> 92.69 | Helped |
Closest papers by meaning
- Explaining Probabilistic Models with Distributional Values (ICML 2024 · similarity 0.86)
- Support Vector-based Estimation of Multilinear Games for Feature Selection and Explanation (AAAI 2025 · similarity 0.85)
- A representation-learning game for classes of prediction tasks (ICLR 2024 · similarity 0.84)
- Pragmatic Feature Preferences: Learning Reward-Relevant Preferences from Human Input (ICML 2024 · similarity 0.84)
- RankSHAP: Shapley Value Based Feature Attributions for Learning to Rank (ICLR 2025 · similarity 0.84)
- Bandits with Preference Feedback: A Stackelberg Game Perspective (NeurIPS 2024 · similarity 0.83)
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.