Research map / Bandits and online learning / Online convex optimization and games
Banker Online Mirror Descent: A Universal Approach for Delayed Online Bandit Learning
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 the Banker-OMD framework (multiple savings plus investment on default action) instead of vanilla OMD for delayed bandits | delayed adversarial MAB · delayed adversarial MAB | regret bound | Helped |
| Tune action scales with delay-aware schedule σ_t=(1/√t + d_t√(ln(D_t+1)/D_t))^{-1} | Banker-OMD · delayed adversarial MAB | regret bound | No clear effect |
| Adding a log-barrier regularizer (Banker-SFLBINF) instead of 1/2-Tsallis for general-sign losses | delayed scale-free MAB · delayed scale-free adversarial MAB with general losses | regret bound | Helped |
| Doubling trick on unknown loss range L with skipping of rounds whose feedback exceeds the estimate (ablation) | Banker-SFTINF · delayed scale-free adversarial MAB | regret bound | Helped |
| Plug BOLO (self-concordant barrier) into Banker-OMD to handle arbitrary delays | adversarial linear bandit · delayed adversarial linear bandit | regret bound | Helped |
| Adding the Banker-OMD scale-free learning rates for non-delayed adversarial MAB (Banker-SFLBINF with D=0) | non-delayed scale-free adversarial MAB · non-delayed scale-free adversarial MAB | regret bound | Helped |
Closest papers by meaning
- Prudent-Banker: No Extra Fees for Baseline Safety in Adversarial Bandits With and Without Delays (NeurIPS 2026 · similarity 0.91)
- Online Nonsubmodular Optimization with Delayed Feedback in the Bandit Setting (AAAI 2025 · similarity 0.90)
- Generalized Linear Bandits: Almost Optimal Regret with One-Pass Update (NeurIPS 2025 · similarity 0.89)
- Adversarial Combinatorial Semi-bandits with Graph Feedback (ICML 2025 · similarity 0.89)
- Improved Best-of-Both-Worlds Regret for Bandits with Delayed Feedback (NeurIPS 2025 · similarity 0.89)
- Regret Bounds for Adversarial Contextual Bandits with General Function Approximation and Delayed Feedback (NeurIPS 2025 · 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.