Research map / Learning theory and algorithms / Prediction and decision losses
Preference Elicitation as Average-Case Sorting
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 sequentially learned Mallows or Plackett–Luce priors instead of the uninformative Impartial Culture prior | Distribution-Aware Insertion Sort · PrefLib datasets and Jester | average queries per voter: 10.1% | Helped |
| Use a learned Mallows prior rather than Impartial Culture | Distribution-Aware Insertion Sort · PrefLib datasets in Table 1 | average queries per voter: 30.98 -> 27.99; 19.85 -> 17.64; 38.95 -> 35.56 | Helped |
| Use a learned Plackett–Luce prior rather than Impartial Culture | Distribution-Aware Insertion Sort · PrefLib datasets in Table 1 | average queries per voter: 7.46 -> 5.85; 10.69 -> 7.92; 10.52 -> 7.90 | Helped |
| Use a Mallows prior with dispersion parameter φ < 1 | Distribution-Aware Insertion Sort · Mallows φ model | expected pairwise queries: O(m) | Helped |
| Use a mixture of k Mallows models with bounded dispersion parameters | Distribution-Aware Insertion Sort · Mixture of Mallows models | expected pairwise queries: log k + O(m) | Helped |
| Use Plackett–Luce weights w = (1, 2, . . . , m) | Distribution-Aware Insertion Sort · Plackett–Luce model | expected pairwise queries: Ω(m log m) | No clear effect |
| Use exponentially decreasing Plackett–Luce weights w = (1, φ, . . . , φ^(m−1)), with φ ∈ (0, 1/2] | Distribution-Aware Insertion Sort · Plackett–Luce model | expected pairwise queries: O(m) | Helped |
Closest papers by meaning
- Optimal Design for Human Preference Elicitation (NeurIPS 2024 · similarity 0.88)
- Active preference learning for ordering items in- and out-of-sample (NeurIPS 2024 · similarity 0.87)
- Preference Learning Algorithms Do Not Learn Preference Rankings (NeurIPS 2024 · similarity 0.86)
- Selective Preference Aggregation (ICML 2025 · similarity 0.86)
- Preference Diffusion for Recommendation (ICLR 2025 · similarity 0.86)
- Eliciting Kemeny Rankings (AAAI 2024 · 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.