Research map / Bandits and online learning / Multi-armed bandits
Data Poisoning Attacks on Stochastic Bandits
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 |
|---|---|---|---|
| Poison offline reward data using the optimization strategy | ε-greedy, UCB, and Thompson Sampling · Offline attacks | attack success: all the attacks are successful | Helped |
| Poison offline reward data using the optimization strategy | ε-greedy · Offline attacks | poisoning effort ratio: less than 10% | Helped |
| Poison offline reward data using the optimization strategy | UCB · Offline attacks | poisoning effort ratio: less than 2% | Helped |
| Poison offline reward data using the optimization strategy | Thompson Sampling · Offline attacks | poisoning effort ratio: less than 5% | Helped |
| Use adaptive reward-gap estimates and confidence bounds to set attacks | ACE · Online attacks | target-arm pulls: T - o(T) | Helped |
| Use adaptive reward-gap estimates and confidence bounds to set attacks | ACE · Online attacks | accumulated attack cost: O(log T) | Helped |
| Use ACE instead of an algorithm-specific attack | ε-greedy · Online attacks | attack cost: slightly worse | Hurt |
| Use ACE instead of an algorithm-specific attack | UCB · Online attacks | attack cost: much larger | Hurt |
Closest papers by meaning
- Adversarial Attacks on Stochastic Bandits (NeurIPS 2018 · similarity 0.93)
- Practical Adversarial Attacks on Stochastic Bandits via Fake Data Injection (NeurIPS 2026 · similarity 0.90)
- Stealthy Adversarial Attacks on Stochastic Multi-Armed Bandits (AAAI 2024 · similarity 0.90)
- Stochastic Bandits Robust to Adversarial Attacks (ICLR 2025 · similarity 0.89)
- Adversarial Attacks on Combinatorial Multi-Armed Bandits (ICML 2024 · similarity 0.87)
- Leveraging (Biased) Information: Multi-armed Bandits with Offline Data (ICML 2024 · similarity 0.86)
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.