Research map / Bandits and online learning / Contextual bandits and Thompson sampling
Combinatorial 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 |
|---|---|---|---|
| replace linear-model combinatorial bandits (CombLinUCB, CombLinTS) with neural CN-UCB and CN-TS | CN-UCB / CN-TS · synthetic top-K, h1 linear, h2 quadratic, h3 cosine scores (d=80) | cumulative regret | Helped |
| use neural score estimator instead of linear on non-linear (quadratic, cosine) score functions | CN-UCB and CN-TS · synthetic h2 and h3 | cumulative regret trend | Helped |
| draw M=10 samples per arm (optimistic sampling) instead of one sample per arm (ablation) | CN-TS · synthetic quadratic h2, d=120 | cumulative regret | Helped |
| multiple samples versus a single sample per arm at small context dimension (ablation) | CN-TS · synthetic quadratic h2, d=40 and 80 | cumulative regret | No clear effect |
| increase context dimension d from 40 to 80 to 120 | CN-UCB, CN-TS · synthetic quadratic h2 | cumulative regret scalability | No clear effect |
Closest papers by meaning
- Neural Combinatorial Clustered Bandits for Recommendation Systems (AAAI 2025 · similarity 0.91)
- Optimal and Efficient Contextual Combinatorial Semi-bandits with General Function Approximation (NeurIPS 2026 · similarity 0.89)
- Adversarial Combinatorial Semi-bandits with Graph Feedback (ICML 2025 · similarity 0.89)
- From Contextual Combinatorial Semi-Bandits to Bandit List Classification: Improved Sample Complexity with Sparse Rewards (NeurIPS 2025 · similarity 0.89)
- Neural Dueling Bandits: Preference-Based Optimization with Human Feedback (ICLR 2025 · similarity 0.88)
- Thompson Sampling For Combinatorial Bandits: Polynomial Regret and Mismatched Sampling Paradox (NeurIPS 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.