Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Feature and Parameter Selection in Stochastic Linear 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 FS-SCB to select among feature maps instead of regret balancing with SquareCB baselines | Feature selection · Synthetic linear bandit | regret | Mixed |
| Use FS-SCB to select among feature maps instead of regret balancing with SquareCB baselines | Feature selection · MNIST | regret | Helped |
| Use PS-OFUL to transfer from parameter models instead of learning with OFUL | Parameter selection · Synthetic linear bandit | regret | Helped |
| Increase the parameter-model ball radius from b=0 toward b=2√d | PS-OFUL · Synthetic linear bandit | regret | Helped |
| Use PS-OFUL to transfer from parameter models instead of individual OFUL | Parameter selection · CIFAR-10 | mean reward | Helped |
| Increase the number of parameter models M | PS-OFUL · CIFAR-10 | mean reward | Helped |
| Reduce the accuracy of the parameter-model centers and radii | PS-OFUL and RB-OFUL · CIFAR-10 | mean reward | Hurt |
| Use PS-OFUL instead of regret balancing with OFUL baselines | Parameter selection · CIFAR-10 | mean reward | Helped |
Closest papers by meaning
- Universal and data-adaptive algorithms for model selection in linear contextual bandits (ICML 2022 · similarity 0.92)
- Linear Bandits with Partially Observable Features (ICML 2025 · similarity 0.90)
- Sparsity-Agnostic Linear Bandits with Adaptive Adversaries (NeurIPS 2024 · similarity 0.90)
- Generalized Linear Bandits with Limited Adaptivity (NeurIPS 2024 · similarity 0.89)
- Ensemble sampling for linear bandits: small ensembles suffice (NeurIPS 2024 · similarity 0.89)
- Geometry-Aware Approaches for Balancing Performance and Theoretical Guarantees in Linear Bandits (ICLR 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.