Research map / Adversarial robustness and security / Attacks on text and graphs
Adversarial Attacks on Online Learning to Rank with Click Feedback
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 |
|---|---|---|---|
| attack only a random K-1 item set plus target (conservative design) instead of attacking all non-target arms (Trivial_1) | PBM attack on PBM-UCB · synthetic and MovieLens | chosen ratio of target | Helped |
| use proposed attack instead of Trivial_{K-1} baseline | PBM-UCB attack · synthetic | cumulative cost: 50% | Helped |
| use proposed attack instead of Trivial_{K-1} baseline | PBM-UCB attack · MovieLens 20M | cumulative cost: 40% | Helped |
| use proposed attack instead of Trivial_{K-1} | PBM-UCB attack · synthetic and real | target chosen ratio | No clear effect |
| use proposed attack instead of Trivial_{K-1} | CascadeUCB attack · synthetic | cumulative cost: about 30% less | Helped |
| use proposed attack instead of Trivial_{K-1} | CascadeUCB attack · MovieLens 20M | cumulative cost: more than 50% | Helped |
| Trivial_1 baseline on CascadeUCB | Trivial_1 · synthetic and real | cumulative cost / chosen ratio | Hurt |
| use conservative-estimation attack instead of modified prior attack | UCB attack on 2-armed bandit · synthetic, mu1=0.85 | cost and target pulls | Helped |
| increase Delta_0 | PBM attack · synthetic | cumulative cost | Helped |
| increase x in mu ~ U(0,x) (larger gaps) | PBM attack · synthetic | cumulative cost | Hurt |
Closest papers by meaning
- Simultaneously Learning Stochastic and Adversarial Bandits under the Position-Based Model (AAAI 2022 · similarity 0.89)
- Online Learning in the Random-Order Model (ICML 2025 · similarity 0.85)
- Bandits with Ranking Feedback (NeurIPS 2024 · similarity 0.85)
- Practical Adversarial Attacks on Stochastic Bandits via Fake Data Injection (NeurIPS 2026 · similarity 0.84)
- Stochastic Bandits Robust to Adversarial Attacks (ICLR 2025 · similarity 0.84)
- Online Matching with Stochastic Rewards: Provable Better Bound via Adversarial Reinforcement Learning (ICML 2024 · 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.