Research map / Bandits and online learning / Multi-armed bandits
Robust Stochastic Bandit Algorithms under Probabilistic Unbounded Adversarial Attack
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 sample medians and periodic diminishing pure exploration in UCB | UCB · Stochastic multi-armed bandit under probabilistic unbounded attacks | pseudo-regret | Helped |
| Use sample medians instead of sample means with ε-greedy exploration | ε-greedy · Stochastic multi-armed bandit under probabilistic unbounded attacks | pseudo-regret | Helped |
| Remove periodic pure exploration while retaining median-based UCB (ablation) | med-E-UCB · Stochastic multi-armed bandit under probabilistic unbounded attacks | regret | Hurt |
| Use the med-E-UCB exploration schedule instead of med-ε-greedy exploration | Median-based bandit algorithms · Synthetic simulations | average regret | Helped |
| Use median-based robust algorithms instead of the compared existing algorithms | Bandit algorithms · Synthetic simulations | average regret | Helped |
| Use median-based algorithms instead of α-trimmed UCB or α-trimmed ε-greedy | Bandit algorithms · Synthetic simulations | average regret | Helped |
| Use median-based algorithms instead of RUCB-MAB and mean-based baselines | Bandit algorithms · Cognitive radio testbed | regret | Helped |
Closest papers by meaning
- Stochastic Bandits Robust to Adversarial Attacks (ICLR 2025 · similarity 0.93)
- Sparsity-Agnostic Linear Bandits with Adaptive Adversaries (NeurIPS 2024 · similarity 0.88)
- Practical Adversarial Attacks on Stochastic Bandits via Fake Data Injection (NeurIPS 2026 · similarity 0.88)
- A Near-optimal, Scalable and Parallelizable Framework for Stochastic Bandits Robust to Adversarial Corruptions and Beyond (NeurIPS 2025 · similarity 0.87)
- Improved Best-of-Both-Worlds Regret for Bandits with Delayed Feedback (NeurIPS 2025 · similarity 0.87)
- Stealthy Adversarial Attacks on Stochastic Multi-Armed Bandits (AAAI 2024 · similarity 0.87)
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.