Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Pure Exploration in Kernel and Neural 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 NeuralEmbedding instead of RAGE/ActionElim | NeuralEmbedding · Real-world datasets | sample complexity | Helped |
| linear methods (RAGE, LinearEmbedding) on nonlinear-reward Yahoo data | LinearEmbedding · Yahoo | success rate: 100% -> 88% | Hurt |
| KernelEmbedding vs LinearEmbedding on Yahoo | KernelEmbedding · Yahoo | success rate: 88% -> 98% | Helped |
| RAGE on Yahoo vs ActionElim | RAGE · Yahoo | success rate: 100% -> 90% | Hurt |
| NeuralEmbedding on MNIST success rate vs baselines | NeuralEmbedding · MNIST | success rate: 98% -> 100% | No clear effect |
| adaptive low-dimensional embedding vs RAGE on synthetic data as dimension grows | Embedding algorithms · Synthetic | sample complexity | Helped |
| increase number of arms for ActionElim | ActionElim · Synthetic | sample complexity | Hurt |
| KernelEmbedding/LinearEmbedding vs RAGE and ActionElim on synthetic | KernelEmbedding/LinearEmbedding · Synthetic | sample complexity | Helped |
| KernelEmbedding and LinearEmbedding vs neural on MNIST | KernelEmbedding · MNIST | sample complexity | No clear effect |
| outliers on Yahoo dataset | LinearEmbedding and RAGE · Yahoo | mean sample complexity | Hurt |
Closest papers by meaning
- Exploration via Feature Perturbation in Contextual Bandits (NeurIPS 2025 · similarity 0.89)
- Infrequent Exploration in Linear Bandits (NeurIPS 2025 · similarity 0.88)
- Geometry Meets Incentives: Sample-Efficient Incentivized Exploration with Linear Contexts (NeurIPS 2025 · similarity 0.87)
- Scalable Exploration via Ensemble++ (NeurIPS 2025 · similarity 0.87)
- The Batch Complexity of Bandit Pure Exploration (ICML 2025 · similarity 0.87)
- Linear Bandits with Partially Observable Features (ICML 2025 · similarity 0.87)
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.