Research map / Adversarial robustness and security / Certified and empirical robustness
Explicit Defense Actions Against Test-Set Attacks
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 computed optimal inspection day instead of no defense action | AR forecaster · ten futures markets | worst-case loss: 78% to >97% reduction | Helped |
| inspect Thursday rather than Wednesday | running example linear predictor · 2-day forecast | worst-case loss: 3.8 -> 3.25 | Helped |
| replace optimal defense by the heuristic of inspecting the day with maximal |theta| | optimal defense · 10,000 synthetic AR models | percent optimal: optimal only ~55% of the time | Hurt |
| replace optimal defense by the heuristic of inspecting the last day | optimal defense · 10,000 synthetic AR models | percent optimal: optimal only ~49% of the time | Hurt |
Closest papers by meaning
- Data Poisoning Attacks against Autoregressive Models (AAAI 2016 · similarity 0.88)
- Optimal Attack and Defense for Reinforcement Learning (AAAI 2024 · similarity 0.85)
- Finding Needles in a Moving Haystack: Prioritizing Alerts with Adversarial Reinforcement Learning (AAAI 2020 · similarity 0.85)
- Dynamic Risk Assessments for Offensive Cybersecurity Agents (NeurIPS 2025 · similarity 0.83)
- Decision Theoretic Foundations for Conformal Prediction: Optimal Uncertainty Quantification for Risk-Averse Agents (ICML 2025 · similarity 0.82)
- Adversaries With Incentives: A Strategic Alternative to Adversarial Robustness (ICLR 2025 · similarity 0.82)
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.