Research on defenses against data poisoning in online learning
An Investigation of Data Poisoning Defenses for Online Learning
Translation in progress — untranslated text is omitted.
A study systematically examines defenses against data poisoning in online learning and, under a strong threat model, analyzes the conditions under which four standard defenses allow or resist fast poisoning. Under a weaker, more realistic threat model, whether an attack succeeds depends on how "easy" the learning problem is.
- Research on data poisoning defenses mostly assumes an offline setting and lacks rigorous analysis; under online learning (e.g., OGD incremental updates), there is a lack of general understanding of when various defenses are effective and how fast attacks are.
- Using OGD logistic regression as the learner, under the semi-online and fully-online threat models, theoretical conditions for three effectiveness regimes—easy/hard/intermediate—are given for four filtering-based defenses: L2-norm, labeling oracle, L2-distance-to-centroid, and Slab, and these are validated on four real datasets.
- Theoretically, the L2-norm defense is always in the easy regime; in experiments, existing attacks are mostly effective in the theoretically predicted easy regime and ineffective in the hard regime; the authors state that Slab has the widest hard regime, and under fully-online the attack causes the highest online misclassification rate on the high-dimensional hard-to-learn task (IMDB).
- Future directions noted by the authors: extending to more complex models such as neural networks, to more attack objectives such as targeted/clean-label/backdoor, and providing defenses with provable guarantees (Conclusion). The experiments used only OGD logistic regression and four binary classification datasets; the labeling oracle defense was not tested experimentally.
Related papers
Both use: IMDB Reviews
- Explainable Federated Learning via Global-Local Attribution AlignmentAnalysis / interpretabilityICML 2026