Research map / Adversarial robustness and security / Attacks on text and graphs
Blind Attacks on Machine Learners
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 |
|---|---|---|---|
| Uniform attack (malicious distribution uniform over support) in informed learner setting | Mean estimation / Le Cam lower bound on minimax risk · Theoretical minimax analysis (mean estimation, informed learner) | effective sample size reduction / minimax risk lower bound | Hurt |
| Mimic attack (malicious distribution equal to the family of true distributions) in blind learner setting | Mean estimation / Le Cam lower bound on minimax risk · Theoretical minimax analysis (mean estimation, blind learner) | effective sample size reduction / minimax risk lower bound | Hurt |
| Mixing in malicious data with alpha<=0.5 (attacker chooses G to make P1 and P2 indistinguishable) | Blind learner Le Cam analysis · Theoretical analysis, two hypotheses | pairwise KL divergence | Hurt |
| Uniform attack in linear regression with fixed design | Linear regression minimax lower bound (informed learner) · Theoretical analysis, orthonormal design | effective sample size | Hurt |
Closest papers by meaning
- Agnostic Learning under Targeted Poisoning: Optimal Rates and the Role of Randomness (NeurIPS 2025 · similarity 0.86)
- Flexible, Efficient, and Stable Adversarial Attacks on Machine Unlearning (ICML 2025 · similarity 0.86)
- Backdoor Attacks via Machine Unlearning (AAAI 2024 · similarity 0.85)
- On the Learnability of Distribution Classes with Adaptive Adversaries (ICML 2025 · similarity 0.85)
- Phantom Transfer: Data Poisoning can Survive Data-Level Defences (NeurIPS 2026 · similarity 0.85)
- Reconstruction Attacks on Machine Unlearning: Simple Models are Vulnerable (NeurIPS 2024 · similarity 0.85)
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.