Research map / Learning theory and algorithms / Sample complexity
Parameter Estimation for Generalized Thurstone Choice Models
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 |
|---|---|---|---|
| Increase comparison set cardinality k for generalized Thurstone models with double-exponential (Luce/Bradley-Terry) noise | Maximum likelihood estimation of generalized Thurstone choice model · Simulation n=10, m=100, uniform random comparison sets | mean square error | Mixed |
| Increase comparison set cardinality k for generalized Thurstone models with uniform noise distribution | Maximum likelihood estimation of generalized Thurstone choice model · Simulation n=10, m=100, uniform random comparison sets | mean square error | Helped |
| Use Gaussian noise (gamma_F,k = O(1/k^eps)) instead of double-exponential or Laplace noise | Generalized Thurstone choice model sample complexity · Theoretical sample complexity for MSE | number of comparisons required m | Helped |
| Use uniform noise distribution (gamma_F,k = Theta(1/k^2)) instead of double-exponential or Laplace noise | Generalized Thurstone choice model sample complexity · Theoretical sample complexity for MSE and classification | number of comparisons required m | Helped |
| Increase comparison set cardinality from 2 for double-exponential/Luce model | Point-score classification of two classes · Theoretical classification sample complexity | number of comparisons required m | No clear effect |
Closest papers by meaning
- Hypothesis Testing for Generalized Thurstone Models (ICML 2025 · similarity 0.90)
- Generalizing while preserving monotonicity in comparison-based preference learning models (NeurIPS 2025 · similarity 0.85)
- Generalized Top-k Mallows Model for Ranked Choices (NeurIPS 2025 · similarity 0.84)
- Learning Parametric Distributions from Samples and Preferences (ICML 2025 · similarity 0.83)
- Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective (ICML 2025 · similarity 0.83)
- Preference learning made easy: Everything should be understood through win rate (ICML 2025 · 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.