Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Top-k eXtreme Contextual Bandits with Arm Hierarchy
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 the arm hierarchy to reduce evaluation to effective arms | IGW-topk on XMC datasets · amazon-3m | inference speedup: 100x | Helped |
| Use the arm hierarchy with b=10 instead of the no-hierarchy proxy with b=400 (ablation) | IGW-topk · eurlex-4k | progressive mean reward: 29% | Helped |
| Use IGW exploration rather than greedy selection | Top-k contextual bandits with the extreme reduction framework · 6 XMC datasets | win/draw/loss: 6W/0D/0L | Helped |
| Use IGW exploration rather than Boltzmann exploration | Top-k contextual bandits with the extreme reduction framework · 6 XMC datasets | win/draw/loss: 4W/1D/1L | Mixed |
| Use IGW exploration rather than ε-greedy exploration | Top-k contextual bandits with the extreme reduction framework · 6 XMC datasets | win/draw/loss: 6W/0D/0L | Helped |
| Use IGW sampling rather than the other sampling strategies | Top-k contextual bandits in the realizable eurlex-4k experiment · eurlex-4k | progressive mean reward | Helped |
Closest papers by meaning
- Gaussian Process Bandits for Top-k Recommendations (NeurIPS 2024 · similarity 0.90)
- A Bayesian Approach for Subset Selection in Contextual Bandits (AAAI 2021 · similarity 0.90)
- Balancing Performance and Costs in Best Arm Identification (NeurIPS 2025 · similarity 0.88)
- On the Minimax Regret for Contextual Linear Bandits and Multi-Armed Bandits with Expert Advice (NeurIPS 2024 · similarity 0.88)
- New Classes of the Greedy-Applicable Arm Feature Distributions in the Sparse Linear Bandit Problem (AAAI 2024 · similarity 0.88)
- High-dimensional Linear Bandits with Knapsacks (ICML 2024 · similarity 0.88)
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.