Research map / Bandits and online learning / Multi-armed bandits
AC-Band: A Combinatorial Bandit-Based Approach to Algorithm Configuration
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 AC-Band with k=2 instead of ICAR | AC-Band · CNFuzzDD, RCW, and Regions200 | CPU time: 72% faster | Helped |
| Use AC-Band with k=2 instead of Hyperband | AC-Band · CNFuzzDD, RCW, and Regions200 | CPU time: 73% faster | Helped |
| Use AC-Band with k=2 instead of ICAR | AC-Band · CNFuzzDD, RCW, and Regions200 | percent gap to best: 7% worse | Hurt |
| Use AC-Band with k=2 instead of Hyperband | AC-Band · CNFuzzDD, RCW, and Regions200 | percent gap to best: 6% worse | Hurt |
| Increase the number of configurations sampled by using smaller α | AC-Band · All datasets | percent gap to best | Helped |
| Increase the number of configurations sampled by using smaller α at fixed k | AC-Band · All datasets | CPU time | No clear effect |
| Use larger k instead of k=2 (ablation) | AC-Band · All datasets | CPU time | Hurt |
| Use larger k instead of k=2 (ablation) | AC-Band · All datasets | wall-clock time | Helped |
| Use Hyperband, which samples more configurations than AC-Band | AC-Band comparison · RCW and Regions200 | percent gap to best | Hurt |
| Use Hyperband instead of AC-Band | AC-Band comparison · CNFuzzDD | percent gap to best | Helped |
| Use Hyperband without capping | Hyperband · CNFuzzDD and RCW | CPU time | Hurt |
Closest papers by meaning
- ImpatientCapsAndRuns: Approximately Optimal Algorithm Configuration from an Infinite Pool (NeurIPS 2020 · similarity 0.86)
- Oracle-Efficient Combinatorial Semi-Bandits (NeurIPS 2025 · similarity 0.86)
- Optimal and Efficient Contextual Combinatorial Semi-bandits with General Function Approximation (NeurIPS 2026 · similarity 0.84)
- Thompson Sampling for Real-Valued Combinatorial Pure Exploration of Multi-Armed Bandit (AAAI 2024 · similarity 0.84)
- Combinatorial Stochastic-Greedy Bandit (AAAI 2024 · similarity 0.84)
- Greedy Algorithms for Structured Bandits: A Sharp Characterization of Asymptotic Success / Failure (NeurIPS 2025 · similarity 0.84)
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.