Research map / Bandits and online learning / Multi-armed bandits
In-Context Learning for Pure Exploration
Alessio Russo, Ryan Welch, Aldo Pacchiano
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 |
|---|---|---|---|
| meta-train ICPE with learned inference network and RL query policy on binary search | ICPE · Probabilistic binary search, K up to 256 | accuracy and max stopping time: min accuracy 1.00; max stop time equals log2 K (8 at K=256) | Improved |
| learn the query policy with RL instead of acting greedily on the posterior (I-DPT) (ablation) | ICPE · Stochastic bandits, fixed confidence delta=0.1 | stopping-time tail and correctness | Improved |
| use learned ICPE vs Track-and-Stop / Top-Two stopping rules | ICPE · Stochastic bandits, fixed confidence | sample complexity vs correctness | Mixed |
| evaluate under shifted bandit distributions | ICPE · Stochastic bandits, distribution shift | correctness and stopping time | No clear effect |
| use ICPE instead of I-IDS on magic-action bandits | ICPE · Bandits with hidden information, delta=0.1 | sample complexity | Improved |
| use ICPE vs uniform and Deep CMAB on MNIST pixel sampling | ICPE · MNIST pixel sampling, N=12 | accuracy | Improved |
Closest papers by meaning
- Can In-Context Learning Support Intrinsic Curiosity? (NeurIPS 2026 · similarity 0.88)
- Infrequent Exploration in Linear Bandits (NeurIPS 2025 · similarity 0.87)
- Active Context Selection Improves Simple Regret in Contextual Bandits (NeurIPS 2026 · similarity 0.86)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.86)
- EVOLvE: Evaluating and Optimizing LLMs For In-Context Exploration (ICML 2025 · similarity 0.85)
- Provably Efficient Exploration in Inverse Constrained Reinforcement Learning (ICML 2025 · similarity 0.85)
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.
