Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Sequential Experimental Design for Transductive Linear Bandits
Tanner Fiez, Lalit Jain, Kevin G. Jamieson, Lillian Ratliff
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 sequential gap-based allocation and eliminate arms that are provably suboptimal | Transductive linear bandits · Transductive linear bandits | sample complexity | Improved |
| Adapt sampling to the active arms rather than use a static XY allocation | Linear bandits: benchmark example · Linear bandits: benchmark example | sample complexity | Improved |
| Allocate measurements using differences between active vectors rather than a G-optimal design on the vectors themselves | RAGE · Uniform Distribution on a Sphere | sample complexity | Improved |
| Use adaptive RAGE instead of static XY allocation | Transductive example · Transductive example | sample complexity | Improved |
| Sample each arm once | LinGapE · Many arms with moderate gaps | sample complexity | Hurt |
| Compute allocations on the measurement vectors rather than on differences between vectors | ALBA · Linear bandits: benchmark example | sample complexity | Hurt |
Closest papers by meaning
- Robust Pure Exploration in Linear Bandits with Limited Budget (ICML 2021 · similarity 0.88)
- Stage-wise Conservative Linear Bandits (NeurIPS 2020 · similarity 0.88)
- Testing the Feasibility of Linear Programs with Bandit Feedback (ICML 2024 · similarity 0.87)
- Regret minimization in Linear Bandits with offline data via extended D-optimal exploration. (NeurIPS 2026 · similarity 0.86)
- Thompson Sampling For Combinatorial Bandits: Polynomial Regret and Mismatched Sampling Paradox (NeurIPS 2024 · similarity 0.86)
- Feasible Action Search for Bandit Linear Programs via Thompson Sampling (ICML 2025 · similarity 0.86)
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.
