Research map / Bandits and online learning / Multi-armed bandits
Partially Observable Multi-Sensor Sequential Change Detection: A Combinatorial Multi-Armed Bandit Approach
Chen Zhang, Steven C.H. Hoi
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 |
|---|---|---|---|
| use CMAB adaptive sampling instead of random sensor selection (ablation) | POMSCD scheme · Synthetic pattern (i), p=100, delta=0.5 | ADD: 195 -> 48.5 | Improved |
| use CMAB adaptive sampling instead of random sensor selection (ablation) | POMSCD scheme · Synthetic pattern (i), p=10, delta=0.5 | ADD: 24.4 -> 6.94 | Improved |
| use full CMAB instead of simplified CMAB(s) (with correlation cross-influence) (ablation) | POMSCD scheme · Synthetic pattern (i), p=10, delta=0.5 | ADD: 9.15 -> 6.94 | Improved |
| use full CMAB instead of simplified CMAB(s) (ablation) | POMSCD scheme · Synthetic pattern (ii), p=100, delta=0.5 | ADD: 63.5 -> 49.3 | Improved |
| use CMAB instead of TRAS baseline | change detection · Synthetic pattern (i), p=100, delta=0.5 | ADD: 91.9 -> 48.5 | Improved |
| use CMAB instead of NAS baseline | change detection · Synthetic pattern (i), p=10, delta=0.5 | ADD: 157 -> 6.94 | Improved |
| use CMAB instead of CMAB(s) (ablation) | POMSCD scheme · SECOM real data, m=5 | ADD: 44.3 -> 28.0 | Improved |
| use CMAB instead of TRAS | change detection · SECOM real data, m=30 | ADD: 40.1 -> 16.4 | Improved |
| increase number of observed sensors m from 3 to 10 | CMAB · Synthetic pattern (i), p=10, delta=0.5 | ADD: 15.2 -> 4.85 | Improved |
| random sampling (RAND) versus baselines TRAS and NAS | POMSCD scheme · Synthetic p=100 | ADD: 195 vs 91.9 (TRAS) at pattern (i), delta=0.5 | Mixed |
Closest papers by meaning
- Bandit Quickest Changepoint Detection (NeurIPS 2021 · similarity 0.88)
- Optimal Online Change Detection via Random Fourier Features (NeurIPS 2025 · similarity 0.85)
- Reducing sequential change detection to sequential estimation (ICML 2024 · similarity 0.85)
- Fixed-Confidence Multiple Change Point Identification under Bandit Feedback (ICML 2025 · similarity 0.84)
- The Sample Complexity of Multiple Change Point Identification under Bandit Feedback (NeurIPS 2026 · similarity 0.83)
- Uncertainty Quantification for Data-Driven Change-Point Learning via Cross-Validation (AAAI 2024 · similarity 0.83)
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.
