Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Optimal Algorithms for Stochastic Contextual Preference Bandits
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 MaxInP or Sta'D instead of RUCB | Contextual preference bandits · Linear score instances, d=10, K=50 | cumulative regret | Helped |
| use MaxInP or Sta'D instead of Self-Sparring | Contextual preference bandits · Linear score instances | cumulative regret | Helped |
| use MaxInP or Sta'D instead of DTS | Contextual preference bandits · Linear score instances | cumulative regret | Helped |
| increase context set size K (ablation) | MaxInP and Sta'D · d=10, T=1500 | final regret | No clear effect |
| increase feature dimension d (ablation) | All algorithms · K=80, T=1500 | final regret | Hurt |
| use Sta'D instead of MaxInP at large d | MaxInP · K=80, T=1500 | final regret | Helped |
| move to non-linear scores with GP-fitted versions of the algorithms | MaxInP and Sta'D · Quadratic, Six-Hump Camel, Gold Stein, d=3, K=50 | cumulative regret | Helped |
| use kernelized Self-Sparring on non-linear scores (ablation) | Self-Sparring · Non-linear score instances | cumulative regret | Helped |
Closest papers by meaning
- Stochastic contextual bandits with graph feedback: from independence number to MAS number (NeurIPS 2024 · similarity 0.90)
- Variance-aware Regret Bounds for Stochastic Contextual Dueling Bandits (ICLR 2024 · similarity 0.89)
- Active Context Selection Improves Simple Regret in Contextual Bandits (NeurIPS 2026 · similarity 0.89)
- From Contextual Combinatorial Semi-Bandits to Bandit List Classification: Improved Sample Complexity with Sparse Rewards (NeurIPS 2025 · similarity 0.89)
- Neural Dueling Bandits: Preference-Based Optimization with Human Feedback (ICLR 2025 · similarity 0.89)
- Contextual Bandits for Unbounded Context Distributions (ICML 2025 · similarity 0.89)
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.