Research map / Bandits and online learning / Multi-armed bandits
Stochastic Multi-Armed Bandits with Control Variates
Arun Verma, Manjesh Kumar Hanawal
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 control variates in the UCB index (UCB-CV) versus UCB1, UCB-V, EUCBV | UCB-CV · Synthetic Instances 1 and 2 (Gaussian) | cumulative regret | Improved |
| compare against EUCBV, the closest baseline, which uses arm elimination | EUCBV · Synthetic Instances 1 and 2 | cumulative regret | Hurt |
| use Thompson Sampling on normal rewards and CVs | Thompson Sampling · Synthetic Instances 1, 2 and 5 | cumulative regret | Hurt |
| increase reward variance (lower correlation between reward and CV), sigma_v^2 in {1.0,1.5,2.0,2.5,3.0} (ablation) | UCB-CV · Instance 5 | cumulative regret | Hurt |
| use Batching instead of UCB-CV for non-normal rewards (ablation) | UCB-CV extensions · Instances 2, 3, 4 | cumulative regret | Hurt |
| use Jackknifing or Splitting instead of UCB-CV (ablation) | UCB-CV extensions · Instance 3 (Gamma, heavy-tailed) | cumulative regret | Improved |
| use Jackknifing or Splitting instead of UCB-CV (ablation) | UCB-CV extensions · Instances 2 and 4 (Normal, Log-normal) | cumulative regret | Hurt |
| add approximation error epsilon to the known CV mean (ablation) | UCB-CV · Instances 1, 2, 5 | cumulative regret | Hurt |
Closest papers by meaning
- Combinatorial Multi-Armed Bandit with General Reward Functions (NeurIPS 2016 · similarity 0.90)
- Thompson Sampling Algorithms for Mean-Variance Bandits (ICML 2020 · similarity 0.89)
- A Closer Look at the Worst-case Behavior of Multi-armed Bandit Algorithms (NeurIPS 2021 · similarity 0.89)
- Optimal Algorithms for Stochastic Multi-Armed Bandits with Heavy Tailed Rewards (NeurIPS 2020 · similarity 0.89)
- Stochastic Rising Bandits (ICML 2022 · similarity 0.89)
- Regret Minimisation in Multi-Armed Bandits Using Bounded Arm Memory (AAAI 2020 · similarity 0.89)
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.
